Natural Selection

The defining characteristic of Darwinism is the mechanism of natural selection.

Natural selection is the theorized process by which members of a given generation (containing some “modification” in their genomes that is morphologically evident) get “selected” to reproduce or not.

The initial emphasis here should be on the term “natural.” The “selection” process is not “guided” or have any plan or purpose “in mind.” Darwin meant “natural” in the sense of “naturalistic” (as we have discussed over the last couple of weeks). This is a straightforward aspect of “natural selection,” and we should all be clear about that basic concept.

So the real issues we will consider now concern the “selection” aspect of the proposed mechanism. Let us begin with a common attack on the concept of natural selection, which derives from Darwin’s expansion of the phrase to the now famous “survival of the fittest.” Darwin came to treat “natural selection” and “survival of the fittest” as two ways of saying the same thing, and he argued that the latter phrase was more explanatory than the shorter “natural selection.” He also preferred the latter phrase because there was nothing in it suggesting an intelligent selection process.

Religionists have leaped on “survival of the fittest” almost from the inception of the phrase! A common and oft-repeated attack is this….

“Survival of the fittest” is a useless tautology

A tautology is typically a statement that appears to be in typical subject/predicate form, but in which the predicate really just is the subject. Contrast these two statements:

1) My dog is black.

2) My rose is a rose.

In philosophical terms, the first statement is synthetic, which means that the predicate, “black,” is not logically contained in the subject. The subject, “my dog,” does not logically imply having a black coloration. My dog could be gold, red, some mix, etc. So, the predicate, “black,” actually provides additional information about “my dog” that you could not know just from the phrase, “my dog,” by itself.

By contrast, statement (2) is a tautology, and it is called analytically true. The predicate tells you nothing informative about the subject. The statement looks the same grammatically as (1), but in (2) the predicate isn’t adding anything not already contained in the subject. So, by simply “analyzing” the subject term, you already know the predicate. Now, consider another statement that is a tautology; it is analytically true:

3) Bachelors are unmarried males.

Again, on the face of it, it looks grammatically like (1). However, like (2), this statement is not using the predicate to actually inform you of anything additional about the subject. The statements (2) and (3) are true by definition, which is the nature of all analytically true statements. Analytically true statements are true by definition and are all tautologies, while analytically false statements are false by contradiction… they are literally a contradiction in terms.

So, the attack against “survival of the fittest” is quite straightforward, once you understand what a tautology is. The critic of evolutionary theory engages with the evolutionist as follows:

Critic: What does “survival of the fittest” even mean? What are you claiming is the mechanism?

Evolutionist: It’s simply and intuitive. Morphological changes, however slight, can be impinged upon by environmental factors, such that even the tiniest changes can make a particular organism more or less likely to die due to those environmental factors.

Critic: Okay, I’m still not clear. Maybe I should be, but I don’t understand what you mean by “fittest.” I don’t think you mean anything like “the absolutely most fit,” because that would mean that most organisms would not survive. And we clearly see that most do.

Evolutionist: Don’t make it more complicated than it is. This is a simple concept. “Fittest” just means “more likely to survive” than another organism. So, genetic modifications resulting in morphological modifications make one organism more likely to survive than another organism. That’s all we mean by “fitness.” It’s all about likelihood of survival.

Critic: But that sounds like you mean that “fitness” and “survival” are synonymous. I mean, how can you know that an organism (or morphological variation) was “more fit” except to discover that the organism survived (to pass along its genetic material) when other organisms did not?

Evolutionist: Sure. “Fitness” is discovered by what survives. Survival is a function of fitness.

Critic: Ahh, I thought so. You are right that the concept is simple! In fact, it’s tautologically simple! I ask what “survival of the fittest” really means, and you tell me: “‘Survival of the fittest’ just means ‘survival of those that survive.'” But that’s so “simple” that it is useless as an explanation! I want to know about this supposed “mechanism,” and you hand me an empty tautology.

And variations of this basic critical theme have permeated the religionist literature about evolutionary theory, with the overarching result that Darwin’s notion of survival of the fittest is cast as an obvious tautology:

4) Those that survive are those that survive. 

Now, on the face of it, this seems like an extremely damaging attack on evolutionary theory! In fact, many intellectual religionists still publicly argue as though this attack is equivalent to the death blow to evolutionary theory itself.


Is “survival of the fittest” really a tautology?

Contemporary evolutionists have  taken this attack seriously and have vigorously responded. This excellent article on TalkOrigins explains why the attack is not actually a problem for the mechanism of “natural selection” or for evolutionary theory.

As this article says, Coulter’s argument asks: “Who are the ‘fittest?’  The ones who survive!,” apparently a variant of “survival of the survivors”, but it is not abstract individuals who survive, but individuals with specific heritable characteristics.  These characteristics cause improved survival in predicable and often testable ways.

So, we want to know about these “characteristics” that “cause improved survival in predictable and testable ways.” We want to understand how the natural selection mechanism actually works, so that we can attempt to come up with a better description of the mechanism than “the survival of those that survive.” Thus, we must get very clear about what “fitness” really means.

“Fitness” must mean more than merely “survival,” and, indeed, evolutionists argue that it does.

From the same TalkOrigins article, we read:

If that [tautological attack] were true, then all differential survival would necessarily be selection.  But we have a name for differential survival that isn’t selection; it is called drift (basically, changes in a population’s gene pool due to chance).  And in fact we can often perform tests that distinguish selection from drift. We couldn’t do that if selection were just “those that survive survive”.

So, the evolutionist responds that “fitness” is not equivalent with “survival” because there are genetic changes in a population that survive when those changes are not “selected” (this “drift” is necessarily genetic change that does not manifest itself morphologically). And this response is consistent with Mayr’s definition of “evolution” we looked at last week: “changes in trait or gene frequency in a population of organisms from one generation to the next.” And if “fitness” is not conceptually equivalent with “survival,” then “survival of the fittest” is not an analytical relation and, hence, no tautology.

Because evolutionists can define “differential survival” as “changes in gene frequency in a population of organisms” (regardless of what causes the changes), they can note that “such changes survive, even though they did not have morphological manifestations that could have been impinged upon by selection.” So, “selection survival” and “differential survival” are both forms of “survival,” and they are different sorts of survival, because “characteristics generated by drift also survive,” and these are not selected. They are passed down at least one generation, yet they were not selected to be passed down. Thus, the mechanism of natural selection is a different mechanism than drift for producing genetic changes in a population. Thus, survival/reproduction are independent concepts from that of strictly natural selection or “survival of the fittest.” Thus, they cannot have a tautological relation. The tautological attack fails.

And throughout the evolutionist literature and argumentation, you will find the same concepts repeated:

  • “Evolution” is best defined as “changes in trait or gene frequency in a population of organisms from one generation to the next.”
  • Many different mechanisms come into play in producing these changes, with natural selection being just one of them.
  • Darwin was really speaking loosely and conveniently when equating “natural selection” with “survival of the fittest,” and religionists have made far too much of this mere terminology.
  • Survival is a function of many things, with selection (due to “fitness”) being just one of them.
  • The process of natural selection is everywhere observed, and even religionists agree that it occurs. Thus, adaptation or micro-evolution is not really a debatable subject. Calling it a tautology is ridiculous, when everybody agrees that it happens!
  • Put enough micro changes together over time (whether selected or not), and you get macro changes.


A big fly in the ointment

The above points sound sensible at first, and it is even true that most creationists agree that micro-evolution does occur. However, if we are really careful with the concepts, then there are still significant problems in the above explanation.

First, the question about what “survival of the fittest” really is, as a mechanism, was not the question of what “evolution” is in total. We ask what “survival of the fittest” means, and we are told that it is “shorthand” for a mechanism or process in which certain morphological “characteristics” tend to survive when others don’t. We ask what mechanism is at play here,  and we are told that we can see it working everywhere around us, and examples like the Peppered Moth are pointed to. But we are not yet satisfied and want to know what the evolutionist is pointing to in such examples. The evolutionist tells that that several mechanisms, including drift and natural selection produce genetic changes in the gene pool of a population. We understand genetic mutation, so drift is no problem; we can clearly see a causal mechanism at work in producing genetic change. But natural selection is supposed to be another mechanism, something other than drift, that causally produces genetic change, and that we do not yet understand.

And examples like the peppered moth do not help us here, because no genetic change in the species is evident in such examples. Such examples are consistent with Mayr’s definition about frequency of trait in a population, but when that population has the same genome over time, it is not an example of “genetic change” in the relevant sense. So, we don’t see what the evolutionist needs us to see: Genuine genetic change in a species resulting from the mechanism of natural selection.

We are told (as in the TalkOrigins article): “Fitness refers to specific characteristics, not some abstract and unknown generality, and so can be tested against the real world. There is nothing tautological about survival of the longest fur, longest tail feathers, or most bark-like coloration.” So, yes, we note that natural selection “works on” morphological traits in a species.

But we question whether or not it actually is those “characteristics” that are “selected” by some (as yet undefined) mechanism. We press the evolutionist that we still do not know what he/she is really pointing to as the mechanism in “natural selection.” Perhaps we are dense, but we don’t take vague pointing as an explanation. Darwinism hangs on this mechanism, and we don’t understand what it is.

And religionists are far too quick to admit that natural selection is the “mechanism” of “adaptation” or “micro-evolution.” By so doing, they agree about a mechanism that remains undefined, and the evolutionist is exactly right to indict them: “While supposing that Coulter, like many creationists, believes that all characteristics are preloaded into the genome by design she also clearly knows that species do change in response to changes in their environment.  But such changes (like beak shape in response to varying circumstances) are precisely the part of evolution that SoF [survival of the fittest] describes! So why are creationists arguing that a position they already accept is a ‘joke’?  Could it be they don’t understand their own argument?” (Emphasis appears in original)

Religionists would do far better to agree that characteristics do change, statistically speaking, in populations over time, consistent with Mayr’s definition. But they should be far slower to admit that there is agreement about the mechanism of change. Both sides of the debate fall into the same trap: Treating correlation as causality. And if natural selection is actually a mechanism of change, then it must play a causal role in the process! That very causal role is what should be closely questioned.

Let’s take the famous Peppered Moth example again (to this day it is the classic in biological textbooks). Here is the agreed-upon sequence: At first there were mostly light-colored moths. But the industrial revolution deposited soot on the tree trunks where the moths lived, thereby making the trunks dark-colored. As a result of the darkened tree trunks, the birds could now clearly see the light-colored moths, while the dark-colored moths has a natural camouflage against predation. So, the birds took advantage of the situation and ate most of the light-colored moths. Thus, there were fewer light-colored moths to reproduce whatever genetics underlay the light coloration into the next generation. So, the next generation produced fewer light-colored moths, while the dark-colored moths reproduced willy-nilly and had more dark-colored-genetically-predisposed offspring. Over time, rinse and repeat, and the dark-colored moths become hugely dominant in the population. So, natural selection selected (note the active verb) the dark-colored moths for greater survival and reproduction, while it selected light-colored moths for high incidence of death and thus far less reproduction. Therefore the dark-colored genetics and morphological trait changed in frequency in the population, which just is evolution.

This story is the “gospel” story agreed upon by even most religionists, as TalkOrigins points out. So, of course we can see the mechanism at work, however you define it! If you think that Darwin’s phrase reduces to a tautology, so be it. The particular phrase is not the mechanism! Just change the phrase to something like: “Survival of the dark-colored moths” or anything context-sensitive like that. “Fitness” acts like a variable: just plug in whatever morphological characteristic you observe is getting selected. In this case, “fitness” just means “dark-colored” rather than “those that survived.” And there is no tautology in this explanation. Now the replacement for the tautology, “Survival of those that survive,” is the non-tautological: “Survival of those with dark coloration.”

Except that none of this is yet an actual explanation. It is all actually a grand, circular argument. A “circular argument” is an argument that, much like a tautology, is uninformative: The premises so “contain” the conclusion, so that the conclusion is not really an informative result of the premise. For example….

Christian: The Bible is the word of God.

Atheist: Really? Well I don’t even believe that God exists. So why would I believe that statement?

Christian: Well, because the Bible says that God exists.

Atheist: Ridiculous! Why should I care about what the Bible claims about God?

Christian: The Bible tells about God, because it is God’s word.

You see, the Christian in this example is just going around in circles. And, sadly, I have actually heard more than one Christian “argue” this way. But evolutionists argue in much the same way. This is a lengthy dialog, which mirrors the sorts of debates that happen in the literature and online forums, but it ultimately reveals the grand circle:


Evolutionist: Natural selection is everywhere we look. There are many examples, such as the case of the Peppered Moth.

Creationist: Okay, well, something happened, I’ll agree. The population certainly did seem to change from light-colored to dark-colored over time. But what is this “selection mechanism,” this general principle or mechanism that you are referring to when you use this specific example of “it,” whatever “it” is?

Evolutionist: Well, it’s simple. A particular trait, such as being dark-colored, became advantageous. Those moths that had the trait were selected to pass the trait along to future generations, while fewer and fewer light-colored moths got through the increased predation to be able to live long enough to pass their trait along.

Creationist: Really? So was the predation of light-colored moths the actual “selection mechanism?”

Evolutionist: Yes. The birds more easily saw the light-colored moths against the dark-colored trees, so the light-colored ones were the ones most eaten by birds. That left mostly dark-colored moths to pass along that trait to future generations. Before the industrial revolution, the reverse was the case with coloration, and the dark-colored moths got hit hardest by predation.

Creationist: Okay, so in this case the predation was the specific instance of this “natural selection” you are talking about. The predation was the mechanism of “selection.” But there were still some light-colored moths in the population, just as prior to the industrial revolution there were still some dark-colored moths in the population. Right?

Evolutionist: Sure, even the easy predation of the light-colored moths did not completely kill them off. There were always a few left to pass along their trait to future generations, just as had been happening with the dark-colored moths earlier.

Creationist: So it seems like you are saying that this “mechanism” is not really causing a change in the species. It only has the power to change trait-frequency in a whole population, considered statistically. But it does not actually produce anything. It can only filter what is already there to be filtered. Is that right?

Evolutionist: Yes, “selection” is a “filter” that in effect “decides” what traits “work” at any given moment. For the Peppered Moth, predation “filtered” the population and “chose” dark color rather than light color. Over time, fewer and fewer light-colored moths made it through the “filter” to pass their trait along to the next generation.

Creationist: But, I think you are missing my point. Again, some light-colored moths always made it through the “filter,” just as some dark-colored moths had made it through the “filter” before the industrial revolution. So, I’m not clear about what the nature of a “filter” is that doesn’t really filter. I mean, think of any other true filter we know of. It has holes of a certain size. It is impossible for any object larger than the hole size to pass through. So, at least you are saying that the “filter” of natural selection is not one with, shall we say, uniform-sized holes. It’s a pretty, shall we say, imperfect filter.

Evolutionist: Exactly! You are getting it now. All of these analogies, such as “filter” and “fittest” and so forth only go so far. They are crude analogies and nothing more. But the mechanism itself, as we all see it actually working in reality, really does work. It is not a perfect, uniform “filter,” but it does indeed filter!

Creationist: So that would be why there could remain apes in the ape lineage, even though natural selection ultimately selected traits that became homo sapiens? Are you saying that some members of the ape population escaped the “filter” and continued on, even though the traits that make up homo sapiens became planet-dominating? Is the idea that the “filter” only minimizes some traits while eventually maximizing others?

Evolutionist: Yes, in some cases. Some traits clearly have been so selected that they became the basis of huge, dramatic speciation splits, such that one population splits off entirely from the original, leaving the original as it was, and the new population goes on to become dominant.

Creationist: I see your grand picture, I think. But I’m still not clear on the details. So, natural selection often (if not usually) leaves a non-selected trait intact in a population. Over time a trait’s statistical dominance might ebb and flow, but both the “presently selected” trait and the original trait remain intact in the population. So, strictly speaking, the “selection” process is really only producing a statistical distribution. It is not really producing a trait. I mean, we cannot know what trait is supposedly “selected” at any given moment. We can only know (past tense) what “selection” has selected by comparing statistical distributions over time. At one moment it might appear that dark coloration is the trait that is ultimately being produced (selected) in the species. But really, we cannot know that, because in a quite brief time (particularly speaking in geological time), we might find a very brief fluctuation in the distribution of that trait, with light coloration being the dominant trait, all things considered, over the grand time span of the Peppered Moth. And the problem is actually harder, because the genome itself continues to include both traits, and the species genome is really unmodified by the supposed mechanism. As we saw with the Peppered Moth, simply remove industrial pollution, and the light-colored trait dominates again, and, again, with no change in the species genome (which, as far as we know, always contained the genetic material to produce both light and dark colored moths).

Evolutionist: I’m not sure I follow you. It seems that you are saying that we can’t tell what is selected at any given moment. We need some lengthy period of time to really detect selection at work?

Creationist: Exactly! For validation of your notion of natural selection, you look at tiny time-slices in which you detect a change in statistical distribution of a particular trait. Light vs. dark colored moths are your favored example. However, we all agree (if we agree about any of this) that when light coloration was statistically dominant, there was still a significant (although minority) proportion of dark-colored moths. And when dark coloration was statistically dominant, there was still a significant (although minority) proportion of light-colored moths. So, selection was not actually “choosing” one trait to be the “species trait.” It was only “preferring” a particular distribution of the trait in that particular time slice. Presumably if conditions changed in another time slice, the light-colored trait could again become statistically dominant, and, in fact, that is precisely what we did observe in a later time slice.

Evolutionist: Okay, sure, I see what you are saying, but I don’t see how that threatens the notion of natural selection. If anything, you are flat-out agreeing with me that the mechanism is real and has actual and even predicable power. Even in tiny time slices, the population is responding to environmental pressures.

Creationist: Not quite so fast. I agree that birds can end up eating more light-colored moths in a given time slice. But I don’t see that as a generally-applicable “mechanism” like you do. In fact, I don’t see anything generalizable in that example. In the moth example, you fixate on one trait and one trait only, because that is the “obvious” one to you. And you see that one trait changing in statistical distribution in one time slice, so all you see is “the environment producing dark-colored moths.” But I don’t see that time slice as being particularly significant, nor do I see the trait-distribution change as particularly significant, especially when the “mechanism” is not really changing the genetic makeup of the species over time. The species genome is being left alone, even in your preferred target time slice. And even you admit that a change in conditions could readily produce a dominance of light-colored moths again, putting the species right back where it was prior to your target time slice, both genetically (statistically speaking) and morphologically, which is, in fact, what we have actually observed. In other words, if (in the grand span of geological time) you happened to miss what turned out to be a very tiny, very brief foray into dark-coloredness on the part of the Peppered Moth, all you would actually observe about the species was its overall light-coloredness.

Evolutionist: But we were here to see this change! That’s the point! We have observed natural selection in action.

Creationist: No, all you have really observed is a change in the statistical distribution of a particular trait in a population of moths. You have not seen the Peppered Moth, as a species, change in the slightest!

Evolutionist: But that statistical change is what we mean by “evolution.” That is the best definition of it.

Creationist: So, really, what you mean by “fitness” is something like “more likely to survive,” in the context of natural selection of course (because I recognize that you theorize other “mechanisms” of change). But that notion of “natural selection” is actually vastly different from “survival of the fittest.”

Evolutionist: Yes, exactly. Surely you read my (Elliott Sober (1984)) article, where I say flat out: “Fitness is a probabilistic disposition (a propensity) to survive and be reproductively successful.  It is connected with actual reproductive success the way a coin’s bias is connected with actually landing heads more often than tails.”

Creationist: Yes, I did read it, as I am well read. My problem is not with the reading but with the understanding. You evolutionists all say somewhat different things in responding to such questions! So, let’s stick with the “probabilistic disposition” idea of “fitness.” I ask the question, “What is selection selecting?” And you answer: “It is selecting fitness, where ‘fitness’ means a ‘probabilistic disposition’ to survive and pass along genetic material.” So, I respond that your definition of what is happening cannot be talking about individuals, because earlier you were talking about statistical distribution, and there is no “statistical distribution” of an individual; you must be talking about statistical distribution relative to a whole population. So, you are deriving something about individual odds of being “fit” relative to the statistical distribution of its (apparently significant) trait in the whole population. Is this right?

Evolutionist: Yes, again you are getting it. This isn’t that hard if you look at what is obvious, which is what most creationists refuse to do. Look, a given trait changes in statistical distribution. This means that any given individual has certain odds of having that trait. And those individuals having that trait share in the “odds game” of survival that is indicated by the whole population. Just so, after the industrial revolution, the light-colored individuals had much higher odds of getting eaten, and we see the same odds reflected in the population as a whole, as the light-colored trait got literally eaten over time.

Creationist: Yes, yes, but it’s this “over time” bit that I’m still stuck on. You are saying that selection is doing something, like there’s this generalizable force or principle or mechanism at work. But I have not yet seen you describe or explain it. I have not see you demonstrate that it causes anything at all. All I hear you now talking about is statistics and probability. But even you will admit that you derive those probabilities from what you observe in a population in a particular time slice. However, those probabilities don’t enable you to predict what “selection” will do in a population in a future time slice. You cannot predict anything (or in any way know until you later interpret events). For example, if you know that a group of coins is slightly weighted toward heads, then you have a trait with predictive power (so you suppose). You say, “Okay, in a given series of otherwise fair flips, we can predict that a higher than 50/50 incidence of tails will be the result in this population.” But notice that you can have any accuracy with that prediction only insofar as you know and have some control over all of the relevant variables. What if it turns out that the flips are all done by an expert flipper who is able to detect and counteract the weighting and thereby produce more heads than predicted? What if you are flipping coins onto a slightly flexible surface that responds to the indentation of the heads side and thereby produces a half-rotated bounce more often than not, such that the weighting toward heads actually produces a disproportional number of heads-up landings? The list of variables goes on and on, and you attempt to “trap” for those by saying things like, “otherwise fair flips.” But you can control the variables only in a totally artificial way and for a limited time slice. In reality, there are literally countless variables, and they are all impinging on the outcome of the statistical distribution. There is far, far more going on in trait-distribution as a generalizable principle than birds eating light-colored moths.

Evolutionist: Again, I’m afraid that your point eludes me. Even if I admit that there is more going on than just birds eating light-colored moths, there is not less than that going on! Birds are indeed eating mostly light-colored moths. And so environmental pressure is indeed affecting trait-distribution, which is all that I am saying that “natural selection” is. And, oddly what I do hear you saying is that the environment does impinge on the statistical distributions over time, and that is all we evolutionists are saying! So, we appear to agree that the environment affects the distribution of particular traits. But that is exactly what we mean by natural selection, indeed by evolution! That is all of the causal mechanism we are trying to describe.

Creationist: You take me to be saying less and granting more than I am. Let’s go back to the Peppered Moth again. You fixate on one trait that you can identify as changing in statistical distribution in a narrow time slice. And you believe that you have a mechanism to explain that change: bird predation, thought of as a particular instance of the general mechanism of natural selection. And you believe that you know what the “selected” trait was (among all of the traits held by the Peppered Moth as a species) because it is the one you see “changing” in statistical distribution over time in the population. Let’s think about the statistical distribution in another way. So, let’s think about the actual genetic change in the population during that time slice. Let’s think about the many, many genetic markers that could have changed in the population during that time, and then let’s think about the genetic markers associated with the coloration trait. Let’s think about the countless environmental aspects that could in principle have been impinging upon the genetics of the moth population during that time slice. Surely you would agree that, considered as a species, the actual genetic change in the moth population was vanishingly tiny even during your target time slice. As a proportion of moth-traits, the one that was “affected” was virtually statistically insignificant, even though that is the one so “obvious” to you. Now, consider that this example is your paradigm example, the “most dramatic” example you have ever discovered! And what you have to show for this example is really no statistically significant genetic change in the moth as a species. Considering the Peppered Moth across all known time slices, what you really have is exactly zero demonstrable change as a species. So, your definition of “evolution” and of “fitness” is carefully “dumbed down” to fit just such paradigm examples, when even in such a “dramatic case” you have to carefully constrain the time slice and fixate on a particular trait in order to “dramatically demonstrate evolution at work.”

Evolutionist: That is a really uncharitable way of casting the situation. Actually, we have always said that any genetic distribution change is statistically significant! And, in fact, we have always said that natural selection “nudges” rather than “determines” the process of lasting genetic change. And the process of genetic change itself is painstakingly slow. So, of course we get excited when we see such a perfect example of the process and one that happens right before our eyes in a very short time slice! The way you are interpreting it, however, is, flatly, ridiculous!

Creationist: You are still not seeing the point, so I’ll put it in the form a question: What is “natural selection” or “survival of the fittest?” Do not tell me, “Well, it is a change in the distribution…, blah, blah, blah.” You are there talking about the effects… what you have observed after a series of events. I am asking what caused the events, what is the causal power of natural selection.

Evolutionist: Okay, fine. Well, in the case of the Peppered Moth, the causal power of natural selection is that it, namely in this case: predation of light-colored moths, caused there to be fewer light-colored moths in each subsequent generation after the industrial revolution. So, natural selection clearly “filtered out” (most of the) light-coloration as a trait in the population. Is that clear enough for you?

Creationist: No, not yet. Here’s why. You are again referring to predation as a feature of the environment, and you are fixating on the trait of coloration. I agree that it seems obvious, but, as you know, appearances are often deceiving. I am trying to get below the appearances to the underlying, reductionistic principles of what is happening. But you refuse to explain the principle. Instead you keep pointing to particular examples. I’m trying to understand what it is that your various examples have in common… what is the causal principle of this supposed selection that you claim is taking place everywhere we look.

Evolutionist: Well the real problem is that you refuse to see the principle in the examples. It’s so simple and obvious that even most creationists don’t deny it! So why are you being so ridiculously nit-picky? The environment impinges upon morphological traits in organisms within a population. That impingement causes certain traits to become more dominant, and we say that those traits are “selected.” There. That is the principle you claim to seek.

Creationist: Okay, so now you are talking “impingement,” which is a really vague term. In the case of the Peppered Moth, you would cite predation as a particular form of “impingement” upon a trait. Right?

Evolutionist: Any term we use for the principle is going to be somewhat vague, because it has to be “broad” enough to cover the variety of particular examples. So, yes, predation in this case was the particular example of “impingement.” Would you deny that predation caused there to be fewer light-colored moths?

Creationist: No, of course I would not deny that. What you are failing to explain, however, is the general causal principle of what mechanism produces a statistically significant change in trait distribution. It is always suspicious when particular “examples” of “statistical significance” are taken to “demonstrate” some general claim. Correlation is not causation. And particular events do not necessarily hang together according to a general principle. So, your (actually very few) examples do not by themselves demonstrate any general principle.

Also, it seems to me that a principle does not have to be “broad” or vague in order to be correct. If anything, my suspicion is that you are casting the “principle” so broadly that it can be made to appear to fit any example. Just like Christians seeing the hand of God at work in everything, you see the hand of natural selection at work in everything. And the problem in this particular example is that you fixate on the “statistical change” of one trait to the exclusion of all others that make up the species. Light-colored is clearly “fit” in one time slice and clearly “not fit” in another. Yet the species doesn’t actually change! You say that “the environment” is what makes one trait “fit” in one time slice and “not fit” in another. But that doesn’t strike me as a principle or mechanism that actually causes anything. That strikes me as just a way of categorizing that we humans impose on events after the fact. It’s just an interpretational scheme, not an actual mechanism. It’s not like natural selection makes anything statistically significant happen, given your latest slate of definitions. It’s like you “see” some “process” after the fact and then “recognize” some “role” in the environment, when in the grand scheme of things no demonstrable change in the species has taken place; the environment didn’t have any statistically significant impingement upon the species.

Evolutionist: That’s the problem with you creationists. You are the ones fixated on this or that time slice and demanding “clarity” when you have none of your own to offer. And if you don’t see “results” in this or that time slice, you claim that “nothing is really happening.” But we evolutionists do have the long, grand picture view. So we do see in the case of the Peppered Moth a small-scale change that is particularly dramatic given how quickly it happened!

Creationist: But what you call a “change” is only statistically significant when you focus in on both the particular time slice and the particular trait in question. In the totality of the Peppered Moth’s time on Earth, and in the totality of the genome of the Peppered Moth, nothing of statistical significance has happened at all. The “big ripples” that seem so obvious to you, because you are so close to them, are really utterly insignificant and quickly die away into statistical “smoothness” when you back away from your example much at all! You have no general mechanism of change here at all, which is the weight that the supposed “mechanism” is supposed to bear.


Such debates can go on and on, and they often do. The evolutionist sees in a  few particular examples some grand explanatory principle, while the creationist sees no grand principle and no mechanism that can bear the weight put upon it. But the real fly in the ointment is the embedded circularity in the evolutionist’s perspectives. It is not easy to catch, which is why the evolutionary perspective is appealing to many.

The evolutionist must explain how one species “branches off” into another species, in short, how speciation events happen. The mutation process (in all of its forms) is a random process. And even the most vociferous evolutionists instantly admit that a purely random process cannot produce the diversity of complex life we see on Earth. Even the simplest genetic code is far, far too rich in ordered complexity to have arisen by any random process.

When an evolutionist, such as the one in our above dialog or on TalkOrigins refers to “other change processes” that are not “selected,” watch for smoke and mirrors! Even the notion of “drift” is just another random mutation process, and randomness does not get the job done. So, natural selection is the necessary “ordering” and “sorting” mechanism the neo-Darwinist theory has to account for how randomness becomes ordered complexity. It is the “Darwinist” part of neo-Darwinism, and it is not dispensable to evolutionary theory.

Thus, we rightly want to know what this all-important mechanism is, how it works, and what are the principles of its operation! We rightly press on evolutionists to produce a principled account, not just some hand waving at “dramatic examples.”

What do we get in response to such questions? We get a definition of evolution that has the necessary “change” built right in, but, oddly, in statistical talk relative to populations. Here is what I mean.

Cast one way, the concept of “evolution” is indeed based upon a circular definition. If “change” is all that “evolution” means, then all of reality is evolution. Look at the argument, which we’ll call the Argument From Change:

1) Reality (for us) is in time.

2) The passage of time just is (and is perceived as) changes in reality.

3) So, reality in time is just the reality of change.

4) “Evolution” just means “change,” and “change” just means “evolution.”


5) Reality itself is evolution.

But that “definition” does not differentiate anything. It does not distinguish evolution from anything else. Indeed, everything is evolution. So the “account” fails to explicate anything about evolution. The problem clearly appears in premise 4, which is supposed to explain something but really doesn’t.

Thus, biological evolutionists are trying to explain “biological evolution,” which is not just synonymous with and definitive of all change or of reality itself! They are trying to differentiate “biological evolution” from other possible accounts of the origin of species, one of the alternatives being creationism. So, the above argument cannot be about “biological evolution.”

We ask, then, what biological evolution is, and we are told: “Simply, it is random mutation filtered by natural selection, resulting in genome variation over time in accordance with the principles of modern genetics.” And we say, “Great! We understand random mutation, and we all agree that it is insufficient to account for the vast array of changes in the organisms that have inhabited and do inhabit this planet. So, tell us more about this ‘natural selection.’ What is this, and how does it work?”

Now the evolutionist is not entitled to simply refer to “change” as the account! That sort of appeal would throw back to the truism that is the above argument, and that explains nothing. So, literally everything in neo-Darwinism ultimately hangs on natural selection. It is the necessary mechanism of change! Evolutionists try to minimize the threat, but the threat is real. If natural selection fails to explicate, then the entire theory is left without its all-important mechanism. “Drift” by itself simply won’t get the job done, and honest evolutionists admit the facts stated in this paragraph.

So, what definitions of “natural selection” have we examined in these pages so far? We have primarily ones that look virtually identical to Mayr’s definition of “evolution” we looked at last week: “changes in trait or gene frequency in a population of organisms from one generation to the next.”

On TalkOrigins we read: “Fitness refers to specific characteristics, not some abstract and unknown generality.”

And famed evolutionist, Elliott Sober, tells us: “Fitness is a probabilistic disposition (a propensity) to survive and be reproductively successful.  It is connected with actual reproductive success the way a coin’s bias is connected with actually landing heads more often than tails.”

A “probabilistic disposition” of “traits” is exactly what the definition of evolution itself is aiming at. The distribution of a trait changes over time, statistically speaking, such that we observe a different distribution of a particular trait in one time slice than we do in another. And if we can dream up (or “obviously observe”) the environment producing such a change of trait-distribution, then we have a “causal mechanism” of change that we call “natural selection.”

So everything hangs on what is supposedly changing! What the evolutionist needs to explain is how one species branches off into another species, which is ordered (not random) and directed change in both the genome and morphology of the “changing” population. We don’t mean “directed” in the sense of purpose, but we do mean “directed” in the sense of something like “complying with the demands of natural selection,” because the whole point to a mechanism like natural selection is that it somehow “directs” the otherwise random mutations into something specified!

This natural selection process produces nothing new on its own. It’s role is not to produce anything new! That is the job of genetic mutation (in all of its forms). Natural selection can only “filter” the morphologies it is presented with from the process of genetic mutation, and then it can only filter via death or significant reduction in the ability to reproduce, which is just a belated form of death (as it is the refusal of the life of the next generation). It is truly the “death filter.” It builds nothing. It only tears down and reduces what can be torn down and reduced.

With these points in mind, reconsider the Peppered Moth “dramatic example.” We ask, “What did natural selection actually change?” And we are told: “The trait-distribution of light vs. dark coloration.”

We ask, “How was that a change in the species? It appears to be only a statistical change in some subset of a population (that was not, by the way, even the entire species), and it did not result in any genome change in even that population.”

We are told: “There doesn’t have to be a genome change in every example! The point is that a morphological change was filtered by the environment, and that’s all we mean by ‘natural selection.'”

So, then, “natural selection” appears to mean only this: The environment puts pressure on some traits and not on others, in some subset of a species’ population, and that pressure produces directed change in the population that makes up the species. Thus, the species does change as a result of environmental impingement, when “species” here just means a statistical account of the trait under consideration.

Ah, yes, but now we are treading dangerously close to the Argument From Change just above! Here is the Natural Selection (NS) version of it:

1) The environment changes over time.

2) A population of organisms exists in the environment.

3) Traits within the population change over time.

4) Traits more or less fit a particular environment at any given time.

5) The environment pressures the population to better fit (regarding those individuals that don’t die before reproducing).

6) The population’s traits change to better fit the environment.

7) The environmental pressure is what we call “natural selection.”


8) Natural selection pressures trait-changes in the population.

Something very close to this argument is what contemporary evolutionists believe. Of course, a full evolutionary follow-up to this argument of what natural selection is would include something like: “Put enough trait changes together over time, and you end up with a new species.” But we are here focusing on what natural selection even is.

So, what is it? It is “environmental pressure” that correlates with a change in trait-distribution within a population. But that just is “change over time,” as we saw in the Argument From Change. There is nothing here more grandiose than that. Let’s consider the premises in turn.

1) Yes, of course the environment changes over time. That is just reality.

2) Yes, of course the population of organisms exists in reality.

3) Yes, traits in a population change over time. A population of organisms is no more “fixed” and static than is reality itself.

4) Yes, traits do more or less “fit” the environment, but an environment is witheringly complex, and the set of traits possessed by most organisms is witheringly complex. So “mapping” the “fit” of a trait to the environment is what is really at issue here as a general principle.

5) This is a truly contentious claim. The environment is what it is, and organisms are what they are. This “pressure” suggests an active principle, and that is what is needed for a mechanism. But the environment is not “actively” seeking to destroy via the death filter. In fact, a vast array of seriously “unfit” morphologies have survived for long, long periods of time. So, the big, big question here really is: Just how “unfit” must a trait be to gain the attention of the death filter in a statistically significant way?

6) The critical word here is “to.” Here is where evolutionists build in the “directed” nature of the change that is supposedly “produced by” natural selection. Yet the huge question is whether any particular trait changes “to” conform to environmental pressures.

7) This premise merely names as “natural selection” the “death filter” that is supposedly the “pressure” of the environment.

8) Ah, the grand conclusion. But, notice that the grand conclusion is really just another way of saying what (5) and (6) are saying. Here is the circularity: (5) and (6) essentially state that natural selection pressures trait-changes in a population, which is precisely what the conclusion says. So, the NS version of the Argument From Change is really no more informative than is the original. It hides its circularity in a bit of verbiage changes. But you have to already believe in premises (5) and (6) before you would be “convinced” to believe in the conclusion.

The reason that so many creationists are ready to accept the notion of adaptation, which is really accepting natural selection as a mechanism of producing change, is that they are confused by what evolutionists are really claiming changes as a result of natural selection.


Definitions matter!

There is no apparent mechanism here, and there is no “adaptation” to the environment on the part of “the species” (thought of merely as a population of organisms sharing some set of traits). And here is why these definitions really matter.

If you accept all of this population-statistics talk as equivalent to genuine change, then you are giving up the very basis for ever defining a species in the first place. Let me explain.

What is a species? Well, it is a population of organisms defined by a set of traits. We will get into this issue heavily in the next two weeks. But for now, please realize that the set of traits defines what a species even is and differentiates that species from another. Evolutionists themselves need there to be some sort of unambiguous account of what a species is, because they must appeal to speciation events in order to demonstrate that evolution in the sense that is relevant to the creation/evolution debate is happening at all!

Conversely, evolutionists also need the “species boundary” to be very, very fungible and flexible. Evolutionists do not see the “tree of life” as made up of disparate species, really. They see only a grand continuum, with forms merging seamlessly into other forms. So, ironically, on the one hand their arguments depend heavily on a very differentiating account of what a species is, while on the other hand they need that account to be fluid and vague enough that it is easy to “get” from one species to the next in the grand continuum.

But everybody agrees that there are real species divides, where organisms from one side of a divide are in no relevant sense “the same thing” as organisms on the other side of the divide. Perhaps the “divide” is an inability of the two populations to productively interbreed. Perhaps (as is often the case with the classic speciation examples) the “divide” amount to little more than a different set of “traits.”

But this very fact matters much to evolutionists’ definition of natural selection!

Let’s again return to the Peppered Moth example. We see a widespread trait-change from dark to light coloration. In this example, that is the only trait-change that evolutionists care about for example purposes. But here they are stuck between a rock and a hard place.

Rock: Evolutionists need the “change” to be a true change in the genome of the species if the change can be an example of the general principle of natural selection’s power to produce ordered complexity and new species.

Hard Place: Evolutionists need the “change” to be merely a statistically-significant distribution change in some subset of the population in order to conform the example to their low-bar definition of what natural selection can “accomplish.” The NS Argument From Change above argues for this “hard place” approach.

If evolutionists are content with “hard place,” then they do not have a mechanism that has ever been demonstrated to “produce” an actual species change.

If evolutionists are determined to have “rock,” then their “most dramatic” example, the Peppered Moth, is really no example at all. Here is why.

With “rock,” evolutionists get actual defining changes in a species; they get the species becoming something genetically different from what it was as a result of environmental pressure. Unfortunately, they have no examples of this ever happening. The Peppered Moth remains a Peppered Moth with the same species genome over the time we have observed. Bacteria that have resistance to a particular drug remain the same “species” of bacteria they were before, and they remain very much bacteria. In short, none of the changed traits turn out to be definitive of the species. They are what we could call “secondary” or “tertiary” traits, as such traits do not affect the organism’s “primary” traits, which are those that differentiate that organism from organisms of another species.

With “hard place,” evolutionists have examples aplenty! But, unfortunately, these are not examples that express natural selection’s power in the evolutionarily relevant way. They are not examples of changes in the species! They are examples merely of statistical distribution, where the “non-selected” traits continue to exist as if natural selection had no power at all. We literally do not see “natural selection at work” in any relevant way.

In a nutshell, a vague recognition of this problem is what motivates creationists to accept “adaptation” yet deny that it can account for full-blown evolution. The problem is that by accepting even “adaptation,” creationists are granting natural selection more explanatory power than it really has.

When you think of “adaptation,” do you imagine just a statistical redistribution of traits, where the “non-adaptive” trait remains in the species’ genome and in a subset of the “pressured” population?

No, the vast majority of people think of “adaptation” as genuine and lasting (genetically based) morphological changes species-wide. And evolutionists are pleased to have you think this way, because they then point to the fact that the vast majority of people believe in “adaptation” and thereby smuggle in the purported, but never yet demonstrated, power of natural selection to substantively change the genome of a species.

So, to sum up, “natural selection” can be thought of in either a deeply circular way or in a way that wields power it has never been demonstrated to possess. Evolutionists want it both ways. They want to define the principle with the circular, “hard place” argument, which has the lowest possible bar the principle can “get over,” and their examples are of this type. But they also want to demonstrate its power to change species, so they speculate that there is some as-yet-undemonstrated mechanism that can produce such change. They cannot have it both ways. The two ways of thinking about natural selection are incompatible. There is a huge gulf between them that evolutionists propose to bridge merely by vagueness, hand-waving, and speculation.

When you really dig into the terms and definitions, the tactics of evolutionists become clear. And as we head into what speciation events are supposed to be, the huge holes in the theory start becoming even more apparent.