What evidence is there for evolution being non-random?

[articulett]

The emotional appeal and the rational need not conflict. Surely there is truth in what you say, but it's an incomplete explanation, at best. Also, there's plenty of emotional appeal on both sides, and there's nothing wrong with that. Indeed, I think that the refusal to use the word "random" is primarily an emotional, not logical, appeal, but more on that later.

So is calling evolution "random".

Irony.

 

[Meadmaker]

Next reason we aren't supposed to use "random".

"It isn't the best description."

If what we were dealing with was a question of how to lead off a classroom discussion, magazine article, or chapter 1 of a book, it would be correct that starting off by saying "evolution is random" would probably be a poor start. However, what we are actually dealing with is someone who comes to us and says that we are too complex to have happened "randomly", or "by chance". It's already in their head, so we don't have a choice about whether we can address it. If we fail to address it, it appears we cannot address it.

Likewise, there is a similar argument that the word is misleading and it might lead to confusion. Once again, this argument fails because they are already using the word, and they already know what they mean by it. As we have seen, there are multiple meanings and it is possible that their meaning accurately describes evolution, or it is possible that it does not accurately describe evolution. You have to find out what they mean by it, and then address the actual meaning that they are using.

If you insist that it is not random, because there is some meaning of "random" that does not fit, you've failed to address their argument, and they might draw the inference that you cannot address the argument.

 

[mijopaalmc]

So is calling evolution "random".

Irony.

Not really. You seem to be describing evolution as "non-random" purely for rhetorical flourish, whereas I and others are describing it as "random" as a matter of intellectual rigor. You seem to be more concerned with what is "inspiring", where as I am more concerned with what is correct.

 

[I less than three logic]

Not really. You seem to be describing evolution as "non-random" purely for rhetorical flourish, whereas I and others are describing it as "random" as a matter of intellectual rigor. You seem to be more concerned with what is "inspiring", where as I am more concerned with what is correct.

As far as I can tell, your definition of random is far to broad to be used to describe anything really. Instead playing equivocation, why not substitute your definition of random for the word random when you're describing it in such a way.

"Evolution is a process that involves some probabilities."

Don't think anyone would have a problem with that. It's just too vague, it doesn't say anything.

 

[articulett]

Not really. You seem to be describing evolution as "non-random" purely for rhetorical flourish, whereas I and others are describing it as "random" as a matter of intellectual rigor. You seem to be more concerned with what is "inspiring", where as I am more concerned with what is correct.

Oopsy, you missed the conversation again. I am referring to the "non-random" aspects of evolution because that was the topic of this thread, remember? I am not describing the entirety of evolution as "non random" either. I have often stated my preference is "random mutation coupled with natural selection"--and then later fill in the details as to how mutations aren't truly random but all sorts of unpredictable things happen to genomes causing less than perfect fidelity and also recombinants. But only the changes or new combinations that have a survival/reproductive advantage on the possessor (or kills off the competition) get passed on preferentially

I think that is as intellectually rigorous and descriptive as anything you have ever said. Moreover, I think nothing that you have said conveys the facts of evolution as well as those you criticize. If it's only correct in your head per your specific definitions, then what good is it?

And you never did tell me what a non-random process is...nor did you give an example. How can you answer mijos question if there is no such thing as a "random process" according to whatever definition you are using to define "evolution" as random. We already know you'd also describe mate selection as random, because you've said so. So what is a non-random process? Is death a random process per your definition? Do some people who wear seat belts die in car accidents while others who don't wear seat belts means that the process in unpredictable and that it fits a non-uniform distribution on a probability chart? Doesn't this make it non-deterministic? Is it relatively random as to who wears seat belts and when? So is the wearing of seat belts to save lives a random process? A stochastic process? What's the evidence for it being a "non random" process? How is seat belt wearing less or more of a random (or stochastic process) than evolution? And if the whole thing makes your head spin, then I've achieved what Behe does with every sentence he writes.

The word "random" and "non random" are only part of this thread because of the question in the op. Moreover--stochastic has come to mean a synonym for random for mijo...and chance or haphazard tends to be the layperson definition. So can you see yet, why we aim for clearer words when describing evolution. Mijo and You really truly sound about as clear and correct as my seatbelt example. You just are not conveying useful information. And mijo asks stupid questions he doesn't want the answer to.

Intellectual rigor, my ass. I think I've heard Kleinman say the same. You fool only yourself.

 

[articulett]

As far as I can tell, your definition of random is far to broad to be used to describe anything really. Instead playing equivocation, why not substitute your definition of random for the word random when you're describing it in such a way.

"Evolution is a process that involves some probabilities."

Don't think anyone would have a problem with that. It's just too vague, it doesn't say anything.

BINGO

 

[Meadmaker]

Don't think anyone would have a problem with that. It's just too vague, it doesn't say anything.

Suppose a creationist approached you and said, "We are too complex to have come about randomly." and you responded, "'Randomly' doesn't really say anything,"

Have you addressed the argument?

 

[articulett]

Evolution, on the other hand, is based in the probability that an individual of a given phenotype will pass on its genes to the next generation. While the relative fitness of the individual does bias its survival, not every fitter individual passes on its genes and not ever less fit individual doesn't. Thus since there are different outcomes for identical individuals, evolution is a by definition a stochastic process.

Seat belt wearing, on the other hand, is based in the probability that an individual wearing a seatbelt will preferentially survive over non-seat belt wearers. While the wearing a seat bell does bias an individual's survival, not every seat belt wearer survives and not every non-seat-belt wearer doesn't. Thus since there are different outcomes for identical individuals, seat belt wearing is by definition a stochastic process.

You are as clear as that. Really.

And evolution is not based "on the probability that an individual of a given phenotype will pass on its genes to the next generation" even though you say so. ..it's based on DNA strands that are successfully copied (or mutated or recombined) versus those that aren't (or aren't anymore). That's it. If it has what it takes to get itself copied the most, it IS the most fit...if it doesn't get copied it just isn't fit. We're not talking phenotypes or anthropomorphic values--we are just talking the basics. It's the same for "fitness" in a computer virus. By the way, computer viruses evolve just as the protection software evolves--an evolutionary arms race not unlike the one going on in a DNA level where the information is coded in base pairs rather than bytes.

 

[Walter Wayne]

I'm waiting for an example of a non-random process and a random process that is not evolution...

If you can't give such examples, then isn't all the rest meaningless. Otherwise aren't you just saying evolution is more like gas flow than some undefined non-random process?

Of course, such examples have been given numerous times. Starting early in this thread, possibly in some of the other recent threads as well.

Non-random
- A ball bouncing of a wall. If you know the angle it hits the wall, you know the angle it comes of the wall.
- The path a thrown ball takes in a gravitational field. We all know the trajectory it takes from an early age.
- The ideal sieve you proposed above, is also non-random. All things that fit pass through, and none that don't fit.

Random examples abound in so many particle experiments. Where an electron hits a detector in a double slit experiment. The path of an alpha particle through a gold-foil in Rutherfords famous experiment.

Now, back to your sieve analogy, and why it is a bad one. Comparing to natural selection does convey the idea of selection as a filter. But in nature reproductive advantage is no guarantee of reproducing more. Heck, a bird with the genes to be a better flyer or hunter may die in the egg before it even has a chance to realize any advantage over its siblings. The sieve analogy succeeds on the image of a "filter" but fails to convey the details of the filter, namely the element of chance that exists in selection.

Walt

Walt

 

[articulett]

Suppose a creationist approached you and said, "We are too complex to have come about randomly." and you responded, "'Randomly' doesn't really say anything,"

Have you addressed the argument?

Um--Dawkins has. And talk origins. And so have I. And many on this thread. You'd certainly want clarification of their meaning of the word random. Is it "haphazard" per multiple creationist quotes I gave you?

Then you can explain natural selection...how it alters species over time. It worked for me. And many others. Can you say the same for anything you have said? I'd say all of these sources have answered that question much better than you've answered mijo's original post question. And with far more intellectual rigor and better explanatory prowess as well. Clarity beats obfuscating pedantry any day.

 

[articulett]

Of course, such examples have been given numerous times. Starting early in this thread, possibly in some of the other recent threads as well.

Non-random
- A ball bouncing of a wall. If you know the angle it hits the wall, you know the angle it comes of the wall.
- The path a thrown ball takes in a gravitational field. We all know the trajectory it takes from an early age.
- The ideal sieve you proposed above, is also non-random. All things that fit pass through, and none that don't fit.

Random examples abound in so many particle experiments. Where an electron hits a detector in a double slit experiment. The path of an alpha particle through a gold-foil in Rutherfords famous experiment.

Now, back to your sieve analogy, and why it is a bad one. Comparing to natural selection does convey the idea of selection as a filter. But in nature reproductive advantage is no guarantee of reproducing more. Heck, a bird with the genes to be a better flyer or hunter may die in the egg before it even has a chance to realize any advantage over its siblings. The sieve analogy succeeds on the image of a "filter" but fails to convey the details of the filter, namely the element of chance that exists in selection.

Walt

Walt

Wrong...because little pieces may not filter through...they may stick together or have stickiness or be on top of big pieces and big pieces may push through or break apart. This is all irrelevant though until someone understands what is meant by a "sieve" or a "culling process" or honing or evolving. Understanding the basic mechanics of a sieve is important before you go into exceptions. Understanding that when we refer to that which is "selected" or that which is the fittest--we are, in fact, referring to that which reproduced successfully in successful increments through generations while others did not, died, out, faded away, amalgamated in part with something else, etc.

And your examples are about priniciples. Natural selection is a principle. In reality people don't accurately predict ball arcs or every basket ball shoot would be a score. In reality there may not be a "perfect circle" that exists...or a "perfect sphere"--but that is so irrelevant to describing these items that I can't think of why even mentioning that would convey anything of value to anyone.

You are truly in your own conversational world.

It's like you want to mention coefficients of friction and buoyancy and air resistance before anyone has grasped the basic acceleration of gravity.
These details and exceptions and so forth don't change the basic formula--they are later added to fill in the details.

Evolution is like gravity in that they are both scientific theories--and they are both facts. But lets not befuddle the equation or basics with gobbledy gook when simplicity comes first. The question was about how evolution is non-random, remember? A ball bouncing off a wall might be non-random in a way that a particular "selective force" is not-- but you are talking about a single event in that example-not a PROCESS--and series of events connected through time...

Walter, Does Meadmaker and Mijo sound clear to you? Do they describe evolution in a way that makes it understandable. Are you guys on the same page? Is Whitey? How is your understanding different that creationist conundrum #4 (the theory of evolution says that life arose and speciated through random chance.) Surely you don't think this is informative?

And a ball bouncing of a wall in your example requires that you know all the inputs--but how can anyone know all the inputs in evolution--all we know is what did survive--what did get passed on-- we can see that in the genomes all around us. We can do genetic counseling and predict better than we ever could--but with so many variables how can anyone clearly predict far into the future. All we can say is that those genomes that are best at copying themselves today will be the most likely representatives in the life forms of tomorrow. That's a fact. That's deterministic. The ones that don't get copied or die out--won't be represented. That is also a fact. As deterministic as your damn single ball bounce example.

And your definition of "fittest" is fitness per Walter's definition. It is not about how fit a particular strand of DNA is at getting into vectors where it can get copied. Being bad tasting can help an organism survive. Being deformed in a way that makes blending in with the environment can help an organism survive. Being part of a gene that is essential to copying DNA will guarantee that section of DNA's survival. Survival of the fittest was not Darwins term--nor Dawkins. To a biologist, the fittest information in a genome is the stuff that is best at getting itself copied (and added onto). The fittest is the stuff that survives the most elimination rounds through time. Often times, the fittest microbes are ones that don't kill their hosts, but keep them lingering so they can pass on more copies of the microbe. Fitness only refers to ability to get passed on better than competitors for the same resources (mates, food, niche, etc.) "Genes to be a better flyer" don't mean anything. That is only "fitter" to Walter. If there are genes that make you a little faster than the food source next to you, those will be preferentially passed on. If faster reflexes mean you live while the less fast die, then genes involved in your faster reflexes get passed on while the slower dead bird's genes do not. Of course faster prey drives a competition for faster or stealthier or trickier predators...which in turn drive the the prey to become faster or more poisonous or grow spines or spit or burrow, etc. (or rather, those that evolve such traits preferentially survive.)

 

[whiteyonthemoon]

In my education I can think of two examples of randomness being involved in the lesson. In one case, it was presented dishonestly in order not to offend people's sensibilities, and in the other case it was taught honestly.

The first case was with regards to orbital structures. The p and d orbitals were represented as figure 8s rotated about the central spindle (imagine cones with the tips touching and hemispheres on the bases of each). This was done (I have to assume) in order not to offend people's sense of the necessity of continuity. However, it doesn't actually represent anything real. If you are to represent electron clouds the way they actually are, the figure shows discontiguous regions of space in which it is more likely to find the electron(imagine a sphere on either side of the nucleus). The figure given was misleading because it suggested that electrons might be found passing through the nucleus of the atom. This leads to misunderstandings not only of electron orbitals, but to a misunderstanding of the nucleus. Also, it misses an opportunity to address quantum physics where it is relevant.

The other case was in examination of permeable partitions and osmolar pressure. The lesson was not simple, but it lead gradually to an important fact about the universe - that through the law of mass action, probabilities can exert PHYSICAL FORCE in the Newtonian sense. This fact is unexpected and amazing, and students that comprehend it really learn something about the nature of the universe.

So, in relation to pedagogy, what are we to do when randomness rears its ugly head? TEACH THE TRUTH. In the end you have to, because just about everything is touched by randomness in one way or another.

I don't expect this to get through to the person it was intended for. It seems she can't understand two sets of people - engineers and ID proponents. So she won't understand me.

Two Tyrannies: the Tyranny of Lies and the Tyranny of Facts.
Yes, it's very hard to come to any agreement on the nature of reality with shameless liars, but it's also hard to come to an agreement on the nature of reality with a room full of honest bright open minded investigators. Educators may have to deal commonly with the Tyranny of Lies, and they may find it hard to believe that there is a Tyranny of Facts, but even for people who make a living of it, establishing the nature of a system can be a difficult business. For me it's hard to imagine someone having to deal directly with people who aren't interested in rational inquiry but claim they are, and intentionally obfuscate, equivocate, etc. to further their agenda. It would warp my mind to be in that place.

 

[mijopaalmc]

It seems that you have missed some of what I have written as well, articulett. I have also claimed that natural selection (which mirrors you seat belt analogy) is a stochastic process because not every fit individual passes on its genes and not every unfit individual fails to. It may be the case, as it is in the case of the seat belts, that a particular combination of genes heavily biases an individual toward passing on its genes or not, just like a seat belt heavily increases one's surivival. Such a case, however, still only means that said individual is only much more likely on average to pass on its genes, just as you are only much more likely on average to to survive a car accident if you wear a seat belt. The point is that you cannot be certain that a specific individual will survive either to pass on its genes or to drive again, but, over a very large number of trials, it becomes clear that a larger number of fitter individuals or individual who wear seat belts survive. Thus, because of this fundamental uncertainty about specific individual cases and relative certainty about a large number of cases, evolution by natural selection or wearing seat belts are governed by statistical (or stochastic) processes.

 

[Walter Wayne]

Wrong...because little pieces may not filter through...they may stick together or have stickiness or be on top of big pieces and big pieces may push through or break apart. This is all irrelevant though until someone understands what is meant by a "sieve" or a "culling process" or honing or evolving. Understanding the basic mechanics of a sieve is important before you go into exceptions. Understanding that when we refer to that which is "selected" or that which is the fittest--we are, in fact, referring to that which reproduced successfully in successful increments through generations while others did not, died, out, faded away, amalgamated in part with something else, etc.

Then your definition of selected is circular. Natural selection being survival of the fittest. The fittest being those that survive. How useless. You are trying to shoehorn this into non-random. I could take something far more variable and uniform and call it non-random using such a circular definition.

And your examples are about priniciples. Natural selection is a principle. In reality people don't accurately predict ball arcs or every basket ball shoot would be a score. In reality there may not be a "perfect circle" that exists...or a "perfect sphere"--but that is so irrelevant to describing these items that I can't think of why even mentioning that would convey anything of value to anyone.

People have predicted ball arcs very well. One of the simplest experiments in Newtonian mechanics is a ball dropped down an angled chute, which leads to a series of suspended hoops a little bigger than the ball. You can repeatedly drop the ball down the chute and it will pass through the series of hoops with out hitting them.

You are truly in your own conversational world.

You accuse others so easily of your own failings. Repeatedly.

It's like you want to mention coefficients of friction and buoyancy and air resistance before anyone has grasped the basic acceleration of gravity.
These details and exceptions and so forth don't change the basic formula--they are later added to fill in the details.

I don't want to mention those. I could just as well have used planetary orbits, where are ability to predict them is limited only by our ability to measure things like distance and velocity accurately. We've been predicting eclipses accurately for sometime.

Evolution is like gravity in that they are both scientific theories--and they are both facts. But lets not befuddle the equation or basics with gobbledy gook when simplicity comes first. The question was about how evolution is non-random, remember? A ball bouncing off a wall might be non-random in a way that a particular "selective force" is not-- but you are talking about a single event here--not a PROCESS--and series of events connected through time...

If you want to talk about selection as a process connected through time, you have to understand it at you have to understand on two levels. First at the lower "event" level. Then at the series of events.

So an individual gene successfully copying itself into the next generation is an event that is random. If it confers an advantage, or is encoded near another gene that confers an advatage it has a greater chance of being continuing. Think of one gene copying itself into the next generation one or more times as the short term "event".

The process by which a gene thrives or fails the long term process is by the amalgamation of these events. Now, when there are a significant number of identical genes in a population we don't observe the randomness, the process is very predictable. For instance, there are a significant number of blue-eyed people around. One blue-eyed person fails to pass on the gene, it has little affect on the human population. The blue-eyed genes will still be spread through the population in the next generation. If suddenly the environment changes to make blue-eyes a great advantage, they will flourish in the population predictably. If they become a great disadvantage the numbers will quickly dwindle in the population. A huge number of random events often produce a very predictable result, like in the gas laws that people have mentioned here.

However, what about small populations? Now chance in those events begins to produce variation over the long term process. Because the future continuance of a gene is based now on only a few events the probability of something different than expect becomes significant.

Those periods where a gene might exist in small numbers do exhibit randomness in that genes that have a favourable selection bias can still die out. Over the long term, those periods can thus have an effect on the process as a whole.

Walt

Out of time tonight.

 

[Ivor the Engineer]

Let’s try this:

The non-random part of evolution (or surviving a car crash) is the bias i.e. shift from zero-mean that natural selection (or wearing a seatbelt) conveys to populations. All real-world populations have a distribution around a mean summarised by a probability density function (pdf). Because of the central limit theorem, the sampling distribution of the mean tends to be Gaussian for large samples from a population with a pdf of any shape.

Under steady-state conditions the mean can be assumed to be fixed. When the environment changes, the mean of the population is shifted through natural selection until a new steady state is achieved.

The simplification articulett and others make for the benefit of explaining evolution is assume the environmental changes stay fixed in one place long enough for a population to become adapted to them. E.g. Species in isolated, stable environments tend to become highly adapted (i.e. little variance around mean) to those environments, while species that have to survive in a wide range of conditions tend to be less well adapted (i.e. wide variance around mean).

For the electronics engineers amongst us, it’s similar to the small-signal approximation we make to avoid complicated mathematics and get a feel for what is going to happen in a system.

By including the messy detail that the environment is sometimes highly variable (i.e. the optimal mean is shifting all over the place), it becomes much harder to explain to someone how evolution ‘works’.

I don’t understand the quest some here have to teach everybody the intricate details of evolution. Do they also think everyone should be taught the fine details of cellular function or relativity? For most people for most subjects the hand-waving argument is just fine.

 

[I less than three logic]

Suppose a creationist approached you and said, "We are too complex to have come about randomly." and you responded, "'Randomly' doesn't really say anything,"

Have you addressed the argument?

Back to equivocation I see. Perhaps you didn't read my whole post? It wasn't very long you know.

"We are too complex to have come about by a process that involves some probability."

I don't know, that would seem to rule out a lot of processes. Doesn't seem like any God created processes would work now either. I'd think most people would agree that God would have the ability to make choices. Thus, He could have chosen not to make humans. So at one point there was both the possibility that humans would and would not exist. That means there was "some probability" involved, therefore we're too complex to have come about that way. Right?

"Too complex to have come about randomly" when you're definition of random is "involves some probability" can only be an argument for hard determinism. I suppose you could now argue about whether God had to make humans or that we had to evolve through natural processes to keep the creationist/evolution thing going if you wanted to.

 

[Meadmaker]

Back to equivocation I see. Perhaps you didn't read my whole post? It wasn't very long you know.

"We are too complex to have come about by a process that involves some probability."

I don't know, that would seem to rule out a lot of processes. Doesn't seem like any God created processes would work now either. I'd think most people would agree that God would have the ability to make choices. Thus, He could have chosen not to make humans. So at one point there was both the possibility that humans would and would not exist. That means there was "some probability" involved, therefore we're too complex to have come about that way. Right?

"Too complex to have come about randomly" when you're definition of random is "involves some probability" can only be an argument for hard determinism. I suppose you could now argue about whether God had to make humans or that we had to evolve through natural processes to keep the creationist/evolution thing going if you wanted to.

I rather like this answer, actually.

The point of my question back to you was to emphasize that this subject is before us not because we are trying to teach evolution, and trying to come up with the best way to do it. We're trying to respond to a creationist argument against evolution.

The answer works, or could work if just slightly rearranged, because it gets at the actual objection. If you find that they mean by "random" nothing more than mijo or I mean, then you can point out that they haven't really said much. That puts them in a position of having to make a more forceful statement if they wish to advance their argument. That more forceful statement then might include a more obvious fallacy, which could be addressed.

If, on the other hand, you deny the existence of the randomness, you can't leave square one. You're arguing against something they know is correct, whether or not they know that it doesn't, by itself, say much.

One thing to remember is that mijo, in this case used as a "randomite" representative, has already said that the random nature of evolution has no particular implications. When asked why he thought it was important, he responded that his main interest was why some people made such a big deal out of it. Implicitly, he is agreeing with your response.

ETA: I, also, agree with your response.

 

[I less than three logic]

I rather like this answer, actually.

The point of my question back to you was to emphasize that this subject is before us not because we are trying to teach evolution, and trying to come up with the best way to do it. We're trying to respond to a creationist argument against evolution.

The answer works, or could work if just slightly rearranged, because it gets at the actual objection. If you find that they mean by "random" nothing more than mijo or I mean, then you can point out that they haven't really said much. That puts them in a position of having to make a more forceful statement if they wish to advance their argument. That more forceful statement then might include a more obvious fallacy, which could be addressed.

If, on the other hand, you deny the existence of the randomness, you can't leave square one. You're arguing against something they know is correct, whether or not they know that it doesn't, by itself, say much.

One thing to remember is that mijo, in this case used as a "randomite" representative, has already said that the random nature of evolution has no particular implications. When asked why he thought it was important, he responded that his main interest was why some people made such a big deal out of it. Implicitly, he is agreeing with your response.

ETA: I, also, agree with your response.

I see your point, and I've never actually said evolution wasn't random myself. I've only asked people define their terms beforehand. I could say "evolution is asterisk", and it could very well be correct depending on how I define the terms I used. The words are not what is important, the meaning they're meant to convey is. Although, in general I'd be against just saying "evolution is random", mainly because I think is probably a bad idea to use such ambitious words when trying to describe something that is often misunderstood to being with.

 

[Meadmaker]

“Selection makes evolution non-random”

This is one of the reasons people say we aren’t supposed to use “chance” when describing evolution and it is said to be one of the counters to use when a creationist says that we are too complex to have arisen by chance.

As a counter-argument, it fails, and it fails for a couple of reasons.

Selection doesn’t produce anything. It just throws out the things that don’t work. It helps select which organisms pass on their genes, but it can’t make new genes. The creationist objection to the theory of evolution isn’t that there’s no way to sort the good organisms from the bad organisms. Their objection is that they don’t think any complex organism could have arisen in the first place. Selection addresses the reason why, once an organism exists, it will continue to live and breed. However, if it can’t exist in the first place, it can’t be selected.

Selection, which is supposedly the solution to the problem, is actually the cornerstone of a very successful argument for the creationist side. Creationists recognize that selection actually makes the problem worse. Far from solving the problem of how we arose by chance, it makes it more difficult to arise by chance by imposing an additional constraint. Selection imposes the additional constraint that every incremental change, which can only come about by chance, must also be beneficial all by itself. It’s not enough to take you closer to some eventual better state, every single state along the way must be continually “up” in the fitness landscape. Creationists exploit this in the argument that is their most effective, that of “irreducible complexity”. (Note: That’s “most effective”, not “most accurate”. It’s a very persuasive argument, despite being wrong.)

The argument from irreducible complexity is that even if you could create a complex organism by random assembly, you would have to do it one of two ways. You could use the “tornado in a junkyard” method, which is so improbable that it could be dismissed, or you could use the gradual, incremental change version. The latter is what, of course, really happens, but they assert that even though the end product of a series of changes is beneficial, the intermediate products, each of which must be the result of a random mutation, must also provide an improvement. They assert that until the final “product” (e.g. an eye, flagellum, or blood clotting mechanism) exists, there’s no benefit to the organism. Therefore, they assert, selection prevents evolution from occurring. (You can see a slight variation of this argument over in the kleinman thread.)

So, using selection as the counterargument actually plays into the creationist hand. You still have to evolve things randomly, so you haven’t solved anything, but those random changes must each be beneficial, all by itself.


There is a way to use selection as a part of an effective counterargument, but it’s a tough sell. It’s correct, so that makes it somewhat easier, but it’s a tough sell. This post is sufficiently long, so perhaps I’ll talk about it some other time.

 

[ImaginalDisc]

Selection, which is supposedly the solution to the problem, is actually the cornerstone of a very successful argument for the creationist side. Creationists recognize that selection actually makes the problem worse. Far from solving the problem of how we arose by chance, it makes it more difficult to arise by chance by imposing an additional constraint. Selection imposes the additional constraint that every incremental change, which can only come about by chance, must also be beneficial all by itself. It’s not enough to take you closer to some eventual better state, every single state along the way must be continually “up” in the fitness landscape. Creationists exploit this in the argument that is their most effective, that of “irreducible complexity”. (Note: That’s “most effective”, not “most accurate”. It’s a very persuasive argument, despite being wrong.)

The argument from irreducible complexity is that even if you could create a complex organism by random assembly, you would have to do it one of two ways. You could use the “tornado in a junkyard” method, which is so improbable that it could be dismissed, or you could use the gradual, incremental change version. The latter is what, of course, really happens, but they assert that even though the end product of a series of changes is beneficial, the intermediate products, each of which must be the result of a random mutation, must also provide an improvement. They assert that until the final “product” (e.g. an eye, flagellum, or blood clotting mechanism) exists, there’s no benefit to the organism. Therefore, they assert, selection prevents evolution from occurring. (You can see a slight variation of this argument over in the kleinman thread.)

So, using selection as the counterargument actually plays into the creationist hand. You still have to evolve things randomly, so you haven’t solved anything, but those random changes must each be beneficial, all by itself.

No.



Your assertion that selection plays into the hands of creationists is based on a faulty assumption: that incremenental changes before the "final" product are useless. Nothing could be further from the truth. I'm mildly nearsighted in one eye, but the poor vision it provides me is substantially better than having an extremely myopic eye, or no eye at all. Even a 20/20 human eye only detects three colors (except fr tetrachromats) and is plainly pitiful when combared to the resolution of a falcon's eye. A stubby wing is better than no wings at all, for flying, and weak claws are better than no claws at all, for predators that claw at their prey. The vast majority of traits which are slected for exist in a broad specturm, with lower utility at one end, and higher utility at the other.

You also cannot forget the cost of developing and maintaining traits. A fleet footed prey organism that is introduced into a habitat without any predators is essentially wasting energy in developing that much speed. Selection would tend to drive the population in the direction of lower speeds. Evolution is not goal oriented.