Look out, ghosts in the machine of evolution: That guy over there has a … a vacuum cleaner!!

Elizabeth Liddle @8:

No, “natural selection” doesn’t necessarily favour reproductive excess.

You don't seem to understand natural selection. Elizabeth Liddle @26:

And statistically, across the next generation, traits that tended to make for fecundity in the previous generation will be best represented in the next, and traist that tend to make for reduced fecundity in the previous generation will be less well represented in the next.

Except when it doesn't. Elizabeth Liddle @8:

But in species who nurture their young, smaller, less frequent litters may be the trait that leads to greater prevalence in the next generation.

Smaller and less frequent than what? Others of the same species leaving more offspring more often? And why complicate it with both litter size and litter frequency? Why not just say that leaving fewer offsping can result in leaving more offspring than those who leave more offspring. 'Cause that's what you're claiming. Oh Please. Show us the math on that one? The only way reduced fecundity would be "selected for" would be if it resulted in increased fecundity compared to the alternative. Call it reproductive excess. Elizabeth @24:

If I have inadvertently muddied the waters, I hope I have now clarified them.

Perhaps muddied waters are advantageous. Why do you insist on complicating such a simple issue? Elizabeth Liddle @26:

living things replicate with variants; some variants replicate better than others.

Let's say that some variants leave more offspring, shall we? Let's call that "reproductive excess." Let's call it "fecundity." So as I stated earlier:

The most fecund variants come to dominate a population because they are the most fecund variants. Period.

Call that "natural selection." See how simple that is? Look at how many words it took you to say what took me so few. Mung

@Elizabeth In regards to fodor's comments I might have missed something, but he essentially treat natural selection as an empty concept in his book 'what darwin got wrong'. -"Yes, it’s a “simple heuristic” – so simple that people try to read more into it than is actually there! But it certainly can’t be “transformed to mean anything”. It only means one thing, which is that traits that bias towards successful reproduction in one generation will be best represented in the next." Maybe I wasn't clear in my explication. What I mean by "transformed to mean anything" is that any variant can be said to have provided reproductive advantage even variants that are contradictory. If anything can be used as an advantageous variant, then we're simply back to sqaure one and a tautology. above

Yes, indeed, above, there are a "plethora of reasons why something might reproduce more", including sheer luck, which is why "drift" is another major factor in allele frequency changes across generations. And I deliberately gave examples of "contradictory" phenotypic traits, because the point of "fitness" in Darwinian terms is not that there is a single "goal" - an ideal phenotype that will win the competition, but that the whole thing is in constant flux, the "fitness landscape" being "high dimensioned", and the evolving population being part of that landscape. But just because something is complicated in practice doesn't mean it is complicated in principle. In principle it's extremely simple. And Jerry Fodor has misunderstood the simple principle if he thinks that "natural selection has no explanatory power because it can simply be used to justify anything being 'selected'. Natural selection isn't used to "justify" anything. It is the logical corollary of the two premises: living things replicate with variants; some variants replicate better than others. Natural selection simply is the result of both those things being true. However, we also need to consider that in typical populations of, say, sexually reproducing animals, traits will be normally distributed, and only slightly correlated. So the fittest individuals (those with the greatest chance of successful reproductions) will be those who inherited the best "cocktail" of alleles across a number of dimensions. And statistically, across the next generation, traits that tended to make for fecundity in the previous generation will be best represented in the next, and traist that tned to make for reduced fecundity in the previous generation will be less well represented in the next. We can model this fairly easily mathematically, although it sounds complicated. But it is all still based on that very simple Darwinian principle. Yes, it's a "simple heuristic" - so simple that people try to read more into it than is actually there! But it certainly can't be "transformed to mean anything". It only means one thing, which is that traits that bias towards successful reproduction in one generation will be best represented in the next. Which is simply, obviously, true! The much more serious issue that people really have with Darwinian principle isn't that (most people take it for granted, it seems, that this is what happens in populations, from year to year, and it's been shown repeatedly both in the lab and field), but with the idea that this constant shifting of allele frequency from generation to generation can possibly be powerful enough to lead to the huge morphological changes we see over time, and the huge morphological differences we see between different lineages once they have ceased to interbreed. That has nothing to do with the principle of natural selection, I don't think, but with the mechanisms by which novel traits are created. Darwin, poor chap, didn't have any idea about that, because it simply wasn't known about in his day. Fortunately we know a great deal more :) Elizabeth Liddle

This is what I a have a problem with Elizabeth. There are a plethora of reasons why something might reproduce more some of which are downright contradictory. On one hand we are told that aggression helps on the other non-aggression helps. On one hand we are told that some veriants propagated due to luck others despite of it. jerry fodor make an interesting observation in his book what darwin got wrong and said that natural selection has no explanatory power because it can simply be used to justify anything being "selected". He also refers to the logical fallacy (intensional fallacy) of the notion of selection as part of the darwinian narrative. Something Mung was explaining yesterday. Honestly, I don't see how this is anything more than a simple heuristic that can be transformed to mean anything. above

Mung and above: No, natural selection has a single meaning, and it is a near-syllogism: When things reproduce with variance, variants that reproduce better will become more prevalent in the population. That is all "Natural selection" means - it refers simply to the truism that each generation is sampling biased towards the best reproducers in the previous generation. There is no other meaning. However, what factors make an individual a good reproducer may hugely vary, even within the same environment. For example, camouflage from predators my help; conversely bright colours to draw the attention of mates may help; copious offspring may help; a few offspring well nurtured may help; reciprocal altruism may help; aggression may help. This isn't a "shell game" - whether something "helps" an individual reproduce or not, is, in principle, simply observable (although in practice, trickier). GWAS studies, for instance, find odds ratios for certain disorders for polymorphisms, but exactly the same thing could be done for number of fertile adult children. So no, "natural selection" is not a "place holder" for anything, any more than "artifical selection" is. But in both cases, what is "selected" may vary - for example, long ears, short nose, curly coat, in the case of artificial selection, or speed, agility, aggression, conspicuousness, in the case of the second. Natural selection, as Darwin conceived it) may have it's problems as an account of population change (stochastic drift, for example, is clearly a hugely important additional factor), but lack of clear operational definition isn't one of them. If I have inadvertently muddied the waters, I hope I have now clarified them. Elizabeth Liddle

Above, If your willing to purchase it you can find Remine's book at the ARN site for a decent price. Its on the top of my must have lists. Probably my next purchase. MedsRex

The book looks really interesting. I’ll pick it up out our university’s library.

My copy is in storage, but I can honestly say that this is a book that I don't mind having an extra copy of, so I've ordered another. iirc, ReMine identifies 4 different versions of "natural selection." I can't at this moment recall what they are. The point is to be able to identify the tautological version, which isn't explanatory (as Ilion points out), and distinguish it from versions which are explanatory, but must therefore be different from the tautological version. A shell game. In order to be testable natural selection must be non-tautological. In order for natural selection to establish the fact of evolution it must be tautological. You just have to train yourself to watch for the switch. I hope you can find a copy of ReMine's book. Mung

PS. The book looks really interesting. I'll pick it up out our university's library. above

@Mung -"Evolutionists will switch between, for example, the tautological version to show how it’s an inescapable fact and some other meaning to show that it’s scientific." Do you mind elaborating a little bit on that? This is precisely what I came looking for yesterday. So the tautological version is basically this: The variants with reproductive advantage will reproduce more. Right? Where does the shell game go from there? above

@ Elizabeth -"No, “natural selection” doesn’t necessarily favour reproductive excess. It can, because sometimes reproductive excess is the trait that leads to greater prevalence in the next generation (as with turtles for instance). But in species who nurture their young, smaller, less frequent litters may be the trait that leads to greater prevalence in the next generation. In yet other species, reproductive excess leads to “boom-bust” population cycles, in which populations explode until they exceed the carrying capacity of the environment and then collapse, leading to oscillating cycles of selection for “reproductive excess” and selection for any trait that maximises survival in scarcity." But the way you're presenting it seems like 'natural selection' can be a place-holder for anything. Highest breed, lowest breed, average breed etc. above

It’s things like this that make me reject what appears to be an empty notion such as ‘natural selection’.

The problem with "natural selection" isn't necessarily that it's an empty notion so much that it's used to mean different things. We see Elizabeth struggling with that fact. What one needs to do is learn the ways it is used and how to recognize when someone is switching between the various meanings. Think of it as a shell game, because that's what it is. Evolutionists will switch between, for example, the tautological version to show how it's an inescapable fact and some other meaning to show that it's scientific. Best expose on this tactic: The Biotic Message Mung

Heh. I wasn't criticizing you. But rather pointing to the circularity of that statement, which you yourself clearly illustrated. In fact, your post #10 is pretty much what I had been thinking about yesterday before I came here and posted after a few weeks/months of absebce. above

Well, sorry. I was striving for meaningless circularity by way of explanation. =P Sheesh. Can't please everyone. Mung

@Ilion That's correct, I did not mean by meaningless = false. And yes, I equivocated explanation with meaning there. Nice catch! :) above

@ Elizabeth -"Unintentional intelligence (e.g. evolutionary processes)" Since when are evolutionary processes intelligent? above

... circularity doesn't make a statement false. Ilion

Above, it's not totally empty of meaning, howevermuch it may be empty of explanation. Ilion

-"The most fecund variants come to dominate a population because they are the most fecund variants." This statement is circular and empty of meaning. It's things like this that make me reject what appears to be an empty notion such as 'natural selection'. above

we seem to agree that “natural selection” doesn’t work as a term of agency...

Nature "selects" is a term of agency to the same extent that nature "chooses" is a term of agency. Do you agree?

...to say that “natural selection selects” is clearly tautological.

Except that's not what I am saying. I'm showing that those who object to my use of the term "choose" and claim that "select" is better because it does not carry the intentional baggage of "choose" would be better off selecting some other battle. To select just is, to choose.

Clearly “it” cannot “choose” in the sense we use that verb in connection with an intelligent agent.

Nor can Nature "select" in the sense we use that verb in connection with an intelligent agent. Nature can no more "select" than nature can "choose." Your objection to my selection of the word "choose" is purely arbitrary. But sure, let's talk about "natural selection."

“Natural selection” is simply a process. Specifically, it’s the process by which the most fecund variants come to dominate a population.

The most fecund variants come to dominate a population because they are the most fecund variants. Period. Why complicate things?

Nonetheless it is a “filter” or “bias” that results in copies of the most fecund in one generation dominating the next.

No filter or bias is required. The "most fecund" by definition, leave more offspring than the "less fecund."

So, with appropriate scare-quotes, we can call it a “selection” process. But if you prefer, you can call it an inbuilt bias of the most fecund.

There is no process worthy of scare quotes. There are some who leave more offspring than others. The reasons and causes are varied. "Natural selection" is code for 'we really haven't a clue what the reasons or causes are, but these obviously left more offspring because they are here." Let's call the reason or cause "natural selection" and be done with it. Not science. Mung

"“Natural selection” is simply a process ..." Even the term 'process' is inherently teleological. Ilion

Mung: we seem to agree that "natural selection" doesn't work as a term of agency - to say that "natural selection selects" is clearly tautological. "Natural selection" is simply a process. Specifically, it's the process by which the most fecund variants come to dominate a population. Clearly "it" cannot "choose" in the sense we use that verb in connection with an intelligent agent. Nonetheless it is a "filter" or "bias" that results in copies of the most fecund in one generation dominating the next. So, with appropriate scare-quotes, we can call it a "selection" process. But if you prefer, you can call it an inbuilt bias of the most fecund. But interestingly, to me, the process itself isn't all that different from the process by which things-with-brains "choose" actions - the most "fecund" of possible options, i.e. the option that receives the most excitatory input from other brain networks is the one that gets executed. But let's leave that to one side. No, "natural selection" doesn't necessarily favour reproductive excess. It can, because sometimes reproductive excess is the trait that leads to greater prevalence in the next generation (as with turtles for instance). But in species who nurture their young, smaller, less frequent litters may be the trait that leads to greater prevalence in the next generation. In yet other species, reproductive excess leads to "boom-bust" population cycles, in which populations explode until they exceed the carrying capacity of the environment and then collapse, leading to oscillating cycles of selection for "reproductive excess" and selection for any trait that maximises survival in scarcity. And yes, a breeder could pick the critters with the most offspring and see what s/he gets, and indeed they do, which is why crop yields go up. But if it were the only criterion, there'd be no guarantee that the produce would be edible to humans. in fact, it would be more likely to end up being most attractive to other species that then pooped out the seeds in situ, and possibly distasteful to species (like ours) whoh ate them and pooped into a sewage system. So they would probably end up with something very new that way. Not "entirely new" of course, because descendents of dogs will always be dogs, just as descendents of mammals will always be mammals,and descendents of tetrapods will always be tetrapods (even if they have no legs, e.g. snakes). Elizabeth Liddle

“Natural selection” is, as is implied by the terms itself, a “natural” (as opposed to human assisted) process by which descendents of organisms best adapted to their environment are best represented in the next generation. The only sense in which this “selection” process differs from the process by which a human farmer does the same thing, is that the human farmer has a specific goal in mind – s/he “intends”. And that is precisely the aspect that Dembski specifically excludes from scientific consideration.

Natural selection chooses the winners and losers. Right? eXcept it doesn't, because it cannot choose. It has no choice in the matter. And since it has no choice, it cannot select. Therefore the term itself, is vacuous. Taking the emphasis off human and putting it on nature does not resolve the underlying issue, which is selection. Unless you turn what it means to select something on it's head, that is. Darwin played a rhetorical game, and it's been played ever since. That doesn't make it any less irrational. And one can certainly argue, and has, that "natural selection" does have a goal, which is to favor reproductive excess. And there's no reason a breeder can't look at his population of breeding critters and pick out the ones leaving the most offspring, getting rid of the rest, wash, rinse, repeat. We have to wonder why it is that breeders haven't realized that they could come up with entirely new sort of creatures by following that simple little rule. Mung

Thanks for your post - I'll check out the link, and we can talk later. Elizabeth Liddle

Elizabeth (#4) Thank you for your post. Just a few quick comments. 1. I see that you have a very long history of thinking about this issue: http://pandasthumb.org/archives/2007/01/the-sad-state-o.html I've only had time to do a quick skim, but I do think you are right about one thing: intentions - and for that matter, intentionality in the philosophical sense of the word - are a vital part of the definition of intelligence. 2. I've been thinkibng on and off about the definition over the last few years. It seems to me that intelligence has several ingredients, all of hwihc are lated: (a) Normativity: the ability to act in accordance with a rule; (b) Goal-directedness: the ability to select an appropriate means to achieve an end; (c) Language: the ability to create a code; (d) Justification: the ability to explain, in some language, why you did what you did. I think Professor Dembski's EF does identify bona fide intelligent agents, in the sense I've defined. See his paper with Marks at http://evoinfo.org/Publications/ConsInfo_NoN.pdf . The point is that even if Nature can achieve the goals via natural selection, it still needs to be pre-loaded with information by some intelligence. Nature might possess (b), but not (a), (c) or (d). Nature behaves in accordance with rules, but does not follow rules as intelligent beings do. To follow rules requires taking them as norms which must be adhered to. That presupposes having the intention to keep the rule, as well as intentionality in a very strong sense of the word: not merely being about something (as bacterial motion might be about food), but also grasping the meaning of the rule. Nature cannot explain why she dos what she does, and Nature cannot create codes. Got to run now. Talk to you later. vjtorley

Well, I don't think the habit is confined to "Darwinists". Dembski, by defining intelligence in terms of the capacity to choose, and by specifically excluding intention from his definition, fails to include natural selection from his definition. "Natural selection" is, as is implied by the terms itself, a "natural" (as opposed to human assisted) process by which descendents of organisms best adapted to their environment are best represented in the next generation. The only sense in which this "selection" process differs from the process by which a human farmer does the same thing, is that the human farmer has a specific goal in mind - s/he "intends". And that is precisely the aspect that Dembski specifically excludes from scientific consideration. tbh, I think that is the flaw at the heart of ID. I think Dembski's approach is absolutely valid as a way of detecting "intelligence" by his own, somewhat idiosyncraic definition. His mistake, IMO, is in thinking that by detecting "intelligence" in this way, he is also detecting "intentions. He seems to think that intentions is the inevitable accompanying property of anything that exhibits "intelligence by his definition, even though his definition does not include it as a criterion. And so his "filter" simply does not differentiate between the products of intentional and unintentional intelligence, and thus does not allow us to infer that a given instance of "specified complexity" was a product of intentional intelligence. Unintentional intelligence (e.g. evolutionary processes) can produce the same signature. So: if ID is to be detected in nature, I think the project has to move on to detecting evidence of intentionin the design of living things, not simply intelligence, and I disagree that this is not a scientific question. Intention is, in principle, possible to detect (although difficult to rule out). Behe is closer, I think,with his notion of Irreducible Complexity. Unfortunately, I think Behe is also wrong, for a whole other set of reasons. Elizabeth Liddle

“choose” and “select” in many contexts are synonyms.

Yet I've seen at least two recent posting here at UD challenging the use of "choose" as implying a mind, or intent, a choice by an intelligent agent, while on the other hand displaying rather the use of the term select, as it if did not carry with it all the same implications as choose. Why is that? Are they just playing semantic games? I wonder what criteria they use to select one term over the other. Words mean what they mean. The problem as I see it is that Darwinists often come along and turn words on their heads without any rational reason for doing so. Select: 1: chosen from a number or group by fitness or preference Selection: 1 : the act or process of selecting : the state of being selected 2 : one that is selected : choice; also: a collection of selected things Mung

An excellent question, Mung. "choose" and "select" in many contexts are synonyms. Interestingly, William Dembski defines "intelligence" as "the power and facility to choose between options" http://www.designinference.com/documents/2000.11.ID_coming_clean.htm but specifically excludes "intention" from the realm of scientific enquiry. So it has always seemed to me, that by Dembski's definition, evolutionary processes are "intelligent". And I would actually agree with him that "specified complexity" is indeed the signature of "intelligence" under his own definition. Although I would disagree with him that "intention" is outside the field of scientific enquiry! It seems to be inside mine :) Elizabeth Liddle

Why is ascribing the power to choose to nature bad, while ascribing the power to select to nature good? Why is one intentional while the other is not? Mung