The Explanatory Power of Darwinism

@jongarvey’s most recent article is a good read, and is likely more correct than most popular work on evolution. He took serious the corrective of The Neutral Theory of Evolution, and it shows here, in a good way.

In that spirit, then, I suggested that epicanthic folds are more likely to be just a near-neutral matter with no biological significance, and all the stories about adapatations to extreme cold (or extreme heat – both have been proposed!) are just that – stories of the Just So variety.

That is likely right. Though, of course, some things will be positively selected. Without more information though, the default null hypothesis should be neutrality.

But the story didn’t end there. George fished out a reference to a key gene for epicanthic folds actually being thought to be an adaptation to increase lactation in low Vitamin D conditions (in an ancient population in Beringia, near the Bering Strait). This invokes pleiotropy, the fact that genes often code for two, or many, entirely disparate functions. …

One might add that there’s no reason that the lactation function might not itself be a spandrel of some third, unguessed, function, or that the whole linked genetic network might consist only of neutral changes fixed by drift. It’s all possible.

Yes it is possible. We because of pleiotropy unexpected things are connected. Because of our amazing abilities, its not even clear what physical characteristics are actually advantageous. After all, we can just reshape our environment at will. It is not clear, for example, Asiatic ethnicities have much true adaptive advantages over African ethnicities for asian.

Yes, there are some good counter examples. E.g. lactose resistance (farming), hemoglobin resistant to malaria (africa), hemoglobin tuned for high altitude (himalayas). These seem, however, to be the exception rather than the rule. Most adaptive features we so adaptive we all got them. Despite our instincts, most “racial” differences are superficial and have little to do with adaptation. We are largely just neutral variations on the same theme.

Note the logic here – even if neutral theory appears overwhelmingly dominant scientifically, adaptive selection must occur because nothing else explains the amazing functionality of the living world.

Well we know that both happen. So this is not that crazy @jongarvey. The exact path of selection and sequence and stepwise function, however, is very difficult to untangle.

However, as the late Austin Hughes also showed:

Contrary to a widespread impression, natural selection does not leave any unambiguous “signature” on the genome, certainly not one that is still detectable after tens or hundreds of millions of years. To biologists schooled in Neo-Darwinian thought processes, it is virtually axiomatic that any adaptive change must have been fixed as a result of natural selection. But it is important to remember that reality can be more complicated than simplistic textbook scenarios.

Exactly right, which makes this a non-sequitur (even though it proceeds :slight_smile:) :

No – if natural selection cannot be demonstrated, then special creation is still available for hire…

This gets to another point. Even when selection is important, we may not be able to detect it. It turns out it is very difficult to detect signatures of moderate to weak selection. So just because something cannot be demonstrated to be selected, does not mean in fact it was not selected.

The demonstration of adaptive selection is indeed problematic at several levels, not least the rarity of truly beneficial mutations, the wildly different estimates up to 1 in a million or more showing that they’re not really measurable. That looks bad for a mechanism that is supposed to account for everything in nature that works well.

Well that is just a math error =). If beneficial mutations are 1:10^6, how many do we expect in a single generation in a population of 10,000, with about 100 mutations per birth? Turns out we expect 1 beneficial mutation per generation. What if there are 1 million people? We’d expect a whopping 100 mutations each generation. Remember, positive selection is going to work to fix these mutations much more rapidly than neutral evolution. Even if half of them are lost, that is quite a lot of beneficial change.

How many non-beneficial mutations were there? Almost a million times more! Do you see the paradox? In one sense, most mutations are neutral. However, that does not mean that beneficial mutations are rare at the population level. Population size serves to amplify the likelihood of coming across beneficial mutations. However, the number of neutral mutations will always totally overwhelm the number of beneficial mutations.

The math here is critical to make sense of this. Do you get it @jongarvey?

Given pleiotropy – with genes coding for sometimes hundred of functions, or even more with alternative splicing – then even if a gene could be demonstrated to be “under selection”, it doesn’t tell you that oriental eyes are advantageous, rather than lactation with less need for Vitamin D, or slightly altered height, or any one or more of the other linked functions. Given that there are those suggesting that every gene has some role in every function, that too is unhelpful in explaining traits.

Very true, though its not really an all-to-all relationship, and down in the weeds and contorted explanations like this might be true, and also the explanations that molecular evolutionary biology is concerned with.

In fact, when science jettisoned universal adaptive natural selection in favour of today’s near-neutral theory, it greatly weakened the observational axion of Darwin that species are wonderfully adapted to their environment. Or rather, it didn’t weaken the observation, but the theoretical basis explaining the obervation, as Joe Felsenstein objected. The guy who understands the theory but doesn’t observe nature may, it’s true, deny the adaptation – but in that case we should criticise his myopia, not doubt our own knowledge of nature.

A better way of putting this is that species may be well adapted to their environment, but this does not explain much of their particulars. It turns out there are several ways to be “well-adapted” and the precise reason we see what we see in a given species might have very little to do with adaptation, even though it very well may be “well adapted.”

Darwinism turns to have very poor explanatory ability for this reason. Neutral theory however, and common descent, do have high explanatory ability. The reason why birds fly with feathers, whales have hair, bats lactate, and human hair goes gray has much more to do with common descent and neutral changes than adaptation.

As every population geneticist with an ounce of competence will tell you, Darwinism is dead and some version of neutral theory reigns uncontested, only refined. This has been the real revolution in evolutionary science.

But under today’s best theory, we just can’t know how the elephant got his trunk. Maybe it was by special creation after all.

That is true too. We don’t quite know.

I should point out though @jongarvey, that there are ways to test hypotheses too. We cannot always resolve what happened, but sometimes we can. There is a way to get at some of these questions. “Just So” adaptationalist stories, however, are to be doubted. A priori, they are usually false, just like mutations a priori are usually neutral.

Thanks Josh - but to be fair I’ve been writing about the ascendancy of neutral theory since 2011. That’s why my thread was entitled “The explanatory power of evolution”, not “…of Darwinism” as in this thread’s title. One does need to keep being reminded, though, when adaptationism still prevails everywhere from David Attenborough to BioLogos.

Of course not - but as an argument, it’s of the same kind as “just because something cannot be demonstrated to be irreducibly complex does not mean it is not.” Lack of demonstration is decreased evidence for the theory. “Just because the action of fairies cannot be demonstrated…”

Felsenstein’s remark was based on assuming what he could not demonstrate, from the functional outcomes. It’s significant that he offered no actual positive evidence for the continuation of adaptive evolution alongside neutral, but simply defaulted to it by the lack of “creative” power in neutral theory.

This is mathematically true, but when one starts looking for specific qualitative examples of beneficial mutations in the literature, they seem to be hiding - and most of those that exist tend to be adaptive through breaking functions that have become counterproductive, rather than by providing new functions. My impression is that the estimates like “1 in a million” are not so much scientific, as faith statements - “they’re so rare we don’t see them, but they must exist, so let’s slap a figure on them.”

The other factor is that, when beneficial mutations are so rare, it’s more risky to regard them as random (even with respect to function). When Fisher assumed a more or less Gaussian distribution around neutrality, benefit was a statistical matter. But when selectable beneficial mutations are an anomaly, it’s much more likely that they ought to be investigated as a class apart, possibly with specific causes directed to adaptation. This, of course, was Asa Gray’s contention - natural selection was just the quality control mechanism for creative innovation.

In that scenario, neutral theory would be the natural noise surrounding the far less common, but truly significant, drivers of evolution. And like Felsenstein, a critic could say that the significance of that is shown by the highly adapted species that surround us, whose wonders don’t show signs of being totally overwhelmed by mildly deleterious mutations.

Well that seems to be a truism, but one with far-reaching implications seldom worked through. Darwin’s starting assumption was the extreme competitiveness of the natural world. What doesn’t adapt, dies. Under neutral theory, the vast majority of changes don’t affect differential survival (exclude purifying selection on highly deleterious mutations here), so dinosaurs do as well with feathers or without, and the struggle for survival is more like a game of croquet.

And that seems to me to leave a hiatus in the explanation for the inexorable progress of living forms over their history. It’s back to the principle of sufficient means, which Darwinism and Creationism seemed to provide, but the current science does not. Which was the theme of my Hump article, of course.

Yup. One can demonstrate that neutral change happens, and that adaptive selection happens, and that placticity-relaxation-mutation happens, and that beneficial variations happen. All testable. But once that overarching explanatory power of Darwinism can no longer be assumed, then which mechanisms have produced what we see, and why, become as open as the causes of history.

That’s great for various versions of creation, of course, since the purpose of the Creator unifies it all. But not for a scientific explanation of living forms.

Of course, only so long as one holds to the preferral of a complete separation of divine causation as an examinable factor within scientific investigation – a distinction to which many do not so irretrievably hold.

Think you might have missed my point, Guy. I did say “various versions of creation,” in all of which the will of God is the unifying principle.

So Darwin (or virtually anybody else) sees the patterns in nature, and in his case particularly superb fit to lifestyle that he called “adaptation.” His theory was a complete explanation of the observed phenomena, rather than just a contributing factor. But now his overarching theory no longer overarches, that observed pattern appears to be the result of:

Some natural selection + neutral drift + pleiotropy + exaptations + niche selection +…

In other words a range of disparate and undirected processes inexplicably makes a coherent pattern. However, add the Creator to the mix, and you get:

(Some natural selection + neutral drift + pleiotropy + exaptations + niche selection +… [+ possible direct intervention])/Divine Purpose

The theological explanation now accounts for a coherent nature including the mechanisms, whereas the mechanisms alone seem more likely to produce confusion than adaptation. A common result would seem to be to squint and conclude that nature is confused, the former impression of order being illusory.

A current example of that is in Ken Miller’s recent book, in which he follows Gould in seeing human intelligence, consciousness and so on as spandrels rather than adaptive features. The language of “accidental by-products” of that approach leaves the most significant human features unexplained, whereas the omniscient Creator doesn’t do accidents, and so the whole concept is fundamentally misleading.

Could be sex selection and/or the result of cultural biases. Hard to tell, though.

Yes we do know how and why the elephant has a long trunk. It wasn’t special creation after all.

1 Like

“Epigenetics makes it possible to be an intellectually fulfilled Lamarckian.”
It’s grain of salt time!

1 Like

Very cool. Looking forward to reading it.

Though, in a theoretical sense, that doesn’t preclude special creation. God still could have directly created that elephant, even though its a thoroughly non-scientific claim.

We don’t know everything there is to know about how elephants got there trunks but we know the most now and that was through hard work by many scientists over a period of years. We will know more and more as more work is done and I am sure we will be both surprised and astonished with the new discoveries as they are reported.

1 Like

A single time special creation is precluded by MN and having a time line of slow changes in trunk sizes.

Well sure. I’m not saying its a scientific theory. Though we’ve covered this before. When I get some time, i’m curious to look at the elephant paper, to see how strong the evidence for positive selection might be.

1 Like

Yeah, this is very weak evidence for positive selection. In fact there appears to be evidence that it is a spandrel. They write:

“Skulls, faces and mouths are formed of interrelated anatomical complexes, and that evolution of one part of these complexes will almost always have an effect on the others,” he says.

About eight million years ago, elephant ancestors relied heavily on grazing from the ground but they had two sets of elongated tusks which prevented them from eating with only their mouths.

The elephant trunk, he says, evolved to such a length to accommodate its large tusks.

So in one aspect of his story Kipling was not so far off after all. The elephant child certainly could eat more with its elongated trunk as can all other elephants alive today.

The story is far more complex than “positive selection for a long trunk” would indicate. Rather it has to do with a complex interrelation between things, that may or may not have been selected for directly. Rather it might be largely the interrelation between parts and drift that gives rise to many of the features we observe.

In regards to the trunk, we have good reason to doubt it is purely some simple story of positive selection. We do not see trunks everywhere. If it was such sharply selected feature, we would see convergent evolution, with other non-elephant trunks (like wings in insects, birds, and bats), but we don’t. That suggest that it might be and adaptive structure (some positive selection for sure), with large components put in place by neutral evolution.

Of course, I could be wrong. The genetics might add more to the story. This, however, is not yet very strong evidence for a Darwinian evolution story. That’s fine though, because adaptationalism alone is not the current model of evolutionary science.

1 Like

This, it seems, was a teleological adaptation - the author appears to be an intellectually fulfilled Lamarckian too.

All we need to explain now is what the two pairs of tusks evolved “for” given that the elephant child had no trunk.

More seriously, if there is an interrelationship between all the features of the facial anatomy, then we may have introduced yet another factor, ie some law of form. You may remember that Stephen Jay Gould debunked the “sexual selection” theory for the antler span of the Irish elk by showing there was a steady relationship between the size of deer and the proportionate span of their antlers.

Yet, as Joshua says, that doesn’t explain why elephants and mastodons, and to a lesser extent tapirs (without the tusks), get trunks, but nothing else does.

And certainly not a reason to invoke a special creation event. Question: Are special creation events instantaneous (of zero time duration)? or could they also be of finite duration? say 1 nanosecond or 8 million years?

1 Like

I totally agree. Though it is also not enough to rule out Special Creation with the appearance of common descent.

1 Like

One of the purposes of discussing all this is to explore possible reasons for a creation event, as well as its nature. So it’s pointless to seek to foreclose the matter.

As for your question, then creation from God’s viewpoint takes place in eternity, and not in time (whether that means instantaneous time or age-long time). But from the material perspective, instaneous, 1 nanosecond or 8 million years are all equally possible.

1 Like

Unless the improvement from the mutation is vast, almost a miraculous improvement one might say, how are they going to get fixed “much more rapidly than neutral evolution” in organisms with tens of thousands of genes? Say the organism had 30,000 genes. If a mutation came along that made a gene’s contribution to fitness a whopping ten times better than its predecessor (an absurdly high number on average absent intervention) then the overall fitness of the organism is improved 1/3000. That is such a trivial amount that I should think it would be lost in the “noise” of life. Maybe it can digest a new food source but it gets eaten by a jaguar, of falls prey to disease, or dies in a fire, before it gets access to said food source and other sources become scarce enough so that the new function matters. It just seems like a lot of things have to go right providentially for even a powerful improvement to fix.

And when I wrote that above, I was only talking about environmental factors which cause the improvement to be “lost in the noise”. The genetic factors are a barrier to fixation that must be got passed before those even matter. What I mean is, harmful mutations probably vastly outnumber beneficial ones. Let’s say 10 to 1. That means for our lucky 11 in a generation who got the positive mutation need to stay lucky on the negative mutations in order for their genome to truly be one that is an overall increase in fitness on net. If they have 10% more negative genes, or their negative genes are 10% more negative than average, then the fitness gain from the positive gene is nullified. If negative mutations outnumber the positive 100 to 1 then the figure is 1%.

The same factors will keep weakly deleterious mutations around, so that I should think true improvement in the kind by chance is difficult indeed.

Since I am sure you are aware of all of this @swamidass and probably know hard data on some of the numbers I am plucking from the air, do you have any good estimation on how big an improvement a new allele needs to fix at a rate significantly higher than that of neutral mutation? Above the noise so to speak.

Patrick I have actually been exploring the gray areas between evolution and special creation/ intelligent design. Reality may be more of a smooth continuum than the boxes and labels we try to put on it.

In science when seeking to change an unknown to a known, invoking a special creation event is considered to be intellectual laziness. It is very lazy to inject miracles and special creation to fill gaps in understanding between what is known and what is still unknown. Outlandish Hypothesis’s and Wild Assed Guesses (WAGs) are okay and even encouraged to get the scientific investigation going on-track or in a completely different direction, but these also would need the hard work of gathering data and facts to support the arguments.

1 Like