More Recent Comments

Wednesday, May 09, 2012

Non-Negotiables of Darwinism

For the sake of completeness, I thought I'd include the basic tenets of Darwinism as outlined by Bill Dembski in his BioLogos essay: Southern Baptist Voices: Is Darwinism Theologically Neutral?.
Non-Negotiables of Darwinism:
  • (D1) Common Descent: All organisms are related by descent with modification from a common ancestor.
  • (D2) Natural Selection: Natural selection operating on random variations is the principal mechanism responsible for biological adaptations.
  • (D3) Human Continuity: Humans are continuous with other animals, exhibiting no fundamental difference in kind but only differences in degree.
  • (D4) Methodological Naturalism: The physical world, for purposes of scientific inquiry, may be assumed to operate by unbroken natural law.
My question is for any scientist who defines themselves as a Darwinist. Do you agree with these non-negotiables? Are there any you would like to add or modify?


58 comments :

Elf Eye said...

How could any of these be considered "non-negotiable"? Aren't the above statements subject to modification should evidence be amassed that is inconsistent with any or all? I thought that that is what science is all about: that any description of reality (theory) is modifiable--even subject to rejection--should it in fact no longer accurately describe reality?

Bryan said...

A few problems there:
D1 - I'd change this to "all known organisms". While unlikely, there may be life on this planet which arose from a unique abiotic process (and somehow went unnoticed all these years). Likewise, any extraterrestrial life that may exist would also be an exception to this rule.

D2 - Natural selection is an evolutionary process, IMO it is not the "principal one", and I think the data supports that.

D3 - I think I agree, although what "difference in kind but only differences in degree" means is a bit unclear. I'd change it though "Humans are continuous with other known life"

D4 - I'd say this rule is true for all sciences - we all assume that the universe operates by set "rules" which are constant and consistent.

Harriet said...

I know that you won't agree with point 2. :)

But yes, "non negotiable" should be replaced by "best evidence indicates" and "2" should read: "Natural Selection is A principle mechanism".

"4": ok, maybe that one is as close to non negotiable as one can get in science.

konrad said...

D1 to D3 are just conclusions based on currently available evidence - we'd discard them if the weight of evidence demanded. D4 really is non-negotiable in the sense that if we discard it we can no longer claim to be doing science.

Ben said...

I don't think #4 is non-negotiable at all. We don't have to assume that the physical world works by unbroken natural law. It has turned out, historically, that everything appears to work that way, but there is no need to assume it. The thing with scientific inquiry is the requirement that evidence be provided for whatever is being postulated. If someone wants to show that something broke natural law, it would take an incredible level of evidence to shake the years of observation we have so far.

K said...

4 is tautological until someone can come up with a nonnatural force that has explanatory power.

Christopher Taylor said...

I would be quite willing to believe, contrary to D3, that humans are qualitatively different from all other animals. This doesn't mean that I'd support Dembski's contrary point. Humans may be different from all other animals, but so are mountain zebras. So are Amazonian giant earthworms. So are Portuguese men-o'-war. So are kokako. So are brindled gnus. So are... well, you get the idea.

SLC said...

I would change D2 to include genetic drift along with natural selection.

steve oberski said...

D4

I'd replace "the physical world" with "reality" as the original wording could be construed to mean that there are some other types of worlds out there.

I'd delete "for purposes of scientific inquiry," as Methodological Naturalism is not limited to any particular sort of inquiry.

I'd replace "unbroken natural law" with "material processes" as Methodological Naturalism is descriptive rather than prescriptive, it's not as if we are telling the universe how to operate, we are discovering the rules of it's operation.


(D4) Methodological Naturalism: Reality may be assumed to operate by material processes.

konrad said...

ps Not that I'd call myself a Darwinist.

pps The wording of D2 is open to a wide range of interpretations. The use of the word "random" to refer to a deterministic but (in practice) unpredictable event is somewhat standard. The natural selection part is more problematic - saying it is "responsible" for adaptations is false in many cases, but it certainly _affects_ all changes (sometimes positively, sometimes negatively, and sometimes with a net effect indistinguishable from zero). So if we replace "responsible for" with "affecting" I might be persuaded to buy "the principal mechanism".

Divalent said...

I agree that NS is the principal mechanism behind biological *adaptation*.

Raul A. Félix de Sousa said...

The whole idea of listing the "non-negotiables of Darwinism" is to try to portray "Darwinism" as a religion with a set of dogmas that could crumble if more scientific evidence is found. But scientific evidence only arises when natural facts are connected to natural causes and, in that sense, there can be no science outside the realm of methodological naturalism. If a natural cause is not found, the problem is unsolved, and that's it. No miracles allowed. As for the other three, there is nothing "non-negotiable" about them. In the case of common descent (D1), all of us "Darwinists" have painlessly accepted the fact that Horizontal Gene Transfer was pervasive in the early stages of life, and that this sentence could be perhaps rephrased substituting "a common ancestor" for "common ancestry". As for D2, the relative importance of natural selection as an evolutionary force has been continuously negotiated over the last fifty years. And finally, in the case of D3, I think Charles Darwin would agree with the "differences in degree", but thanks to scientific evidence, that has been negotiated too, and today most scientists would prefer "Humans are continuous with other animals", period.

Hawks said...

(D1) Common Descent: All organisms are related by descent with modification from a common ancestor.

Darwin once wrote:
There is a grandeur in this view of life, with its several powers, having been originally breathed into a few forms or into one.

So.... methinks that Dembski needs to strike that one off the list. There is NOTHING about evolution that says that life would only be allowed to start once.

(D4) Methodological Naturalism: The physical world, for purposes of scientific inquiry, may be assumed to operate by unbroken natural law.

I suppose that Sober would disagree...

Joe Felsenstein said...

I'd say (D1) is a conclusion, not an assumption. Darwin hedged on this, and even up to the 1950s scientists were not sure that all life came from the same origin. After the details of DNA and its replication were understood there was really no ambuguity.

(D2) is more OK than it looks. He said "adaptations" not "evolution". Of course you can argue that (say) genetic drift causes more evolution than other forces (in that it jiggles gene frequencies back and forth an awful lot). But once you ask about the nonrandomly high degree of adaptation living organisms exhibit, that must be explained by natural selection. Which is different from concluding that any particular adaptation must come about by natural selection. "Principal mechanism" is vague -- it would have been better to talk about natural selection being responsible for the nonrandomly high degree of adaptation. Notice I didn't claim perfect adaptation, just that they are an awful lot better adapted than they would be if you, say, simultaneously mutated every base in the genome in one individual.

In (D3) "no fundamental difference in kind" is vague. (D1) implies continuity already. I fail to see why (D3) is necessary.

(D4) is an OK description of methodological naturalism.

Peter said...

D4 is the only premise required: the others are observed/deduced from the evidence. D2 is OK so long as we're talking about adaptation, however most evolution is non-adaptive through drift.

alias Ernest Major said...

D1: universal common descent is a contigent fact; it is not something that has to be a priori true.

D2: how one slices the world into evolutionary mechanisms is to a degree arbitrary, but random variation and natural selection are two mechanisms, not one. Even if one folds mutation, recombination, inter-population gene flow, inter-specific introgression, etc into variation Dembski has still omitted genetic drift. While natural selection is probably the predominant proximate cause of adaptation it's rather harder to say that the roles of cladogenesis and genetic drift in providing a substrate for exaptation can be neglected.

Random is also potentially problematic. Neither mutation, nor recombination, nor gene-flow, is unbiased. A common formulation is that mutation occur independently of their fitness. I'm not sure that is true; at least at the margins - mutations occur through multiple mechanisms, and some mechanisms may have different probabilities of producing detrimental, neutral or beneficial mutations than others. One might expect that those mechanisms with a greater chance of producing a detrimental mutation are preferentially suppressed by DNA copying/repair mechanisms.

A greater mutation rate in stressed bacteria is in a sense non-random (in comparison to other times and places), and has as a concomitant a greater rate of beneficial mutations.

D3: Dembski seems to confusing Darwinism with atheism. To make the statement true he would need to make it biologically continuous and and biological differences. (By my standards biology is silent on such things as souls. The theory of evolution doesn't care whether God added a soul somewhere along the line.)

D4: Dembski seems to confusing Darwinism with science as a whole. Methodological naturalism is a convenient term to oppose to philosophical naturalism. However, as the distinction between the natural and supernatural is ill-defined, I prefer a formulation that science can only consider processes which display, at least statistically, regularities.

Larry Moran said...

Divalent, I think you're on safe grounds there, but there are other possibilities. There are some very smart biologists (Michael Lynch, Eugene Koonin) who are promoting a nonadaptive theory of the evolution of complexity. As Lynch says in The origins of Genome Architecture (page 389) ...

It is now clear that many (if not most) aspects of genomic biology that superficially appear to have adaptive roots—including the numerous features that contribute to compexity, modularity, robustness, and evolvability—are almost certainly also products of nonadaptive processes, whereas others are readily understood only by invoking a near-complete relaxation of natural selection.

So far, this isn't a challenge to the non-negotiable statement from Dembski but what if further work in this area shows that nonadaptive processes are responsible for a great deal of increased fitness? Would that mean that "Darwinism," or evolution, has been falsified?

Is it a fundamental part of modern evolutionary theory that all adaptations have to be due to natural selection?

Larry Moran said...

I disagree about D4. As a scientist, I feel quite free to investigate the possibility of non-natural processes and I can imagine situations where science could confirm miracles—events that could only be explained by violations of natural law.

Therefore, I don't feel bound by the restriction Dembski imposes. The restriction of methodological naturalism is a convenient way of preventing science from challenging religion and I'm not buying it.

Larry Moran said...

I agree. Thus, none of the so-called "non-negotiables" are valid. Dembski is batting zero. This is actually a pretty good batting average for an IDiot since most of them are below zero.

Larry Moran said...

I prefer a formulation that science can only consider processes which display, at least statistically, regularities.

Does that mean that the idea of a mass extinction caused by a meteor impact isn't science? Does that mean that scientists can't investigate other unique historical events such as the origin of Homo sapiens?

konrad said...

I think your definition of "natural" is too narrow. Many things considered natural today (anything using electricity, for example) would have been considered miracles a thousand years ago. If a supposed miracle were confirmed scientifically, we wouldn't say "hah, science also covers supernatural events", we would say "something previously described as supernatural has turned out to be natural after all, let's set about trying to explain this phenomenon". When we find exceptions to known natural law, we don't say "this is an example of a law of nature being broken", instead we say "we need to generalise our previous formulation of this law, so as to take the newly discovered exception into account". This is a key difference between science and fundamentalist religion.

Chas Peterson said...

Naturally it all depends on your definition of 'adaptation'. Some would argue that "nonadaptive processes are responsible for a great deal of increased fitness" is incoherent, because a trait conferring an increase in fitness (in a given environment) is, to them, an adaptation by definition. Others define 'adaptation' as a trait that evolved "due to natural selection," making Dembski's formulation a tautology. Still others want to call any feature of demonstrated current utility an adaptation mo matter by what process it evolved.

There is a large and contentious literature on the topic. Believe it or not, many actual evolutionary biologists are not willing to give Steve Gould the last word.

Chas Peterson said...

"Amazonian giant earthworms"

I have held one. They feel much more like a squirmy kielbasa than other animals.

Chas Peterson said...

read it again.

Larry Moran said...

So there are three choices?

1. The only possible mechanism of adaptation is natural selection therefore D2 is untrue because the words "principle mechanism" should be "only possible mechanism."

2. There are several possible mechanisms of adaptation but D2 is correct because Darwinism will be falsified if natural selection isn't the "principle mechanism."

3. There are several possible mechanisms of evolution but D2 is incorrect because Darwinism will not be falsified if one of them turns out to be most common than natural selection.

Chas, which one do you agree with? Or is there another option?

Remember that we're talking about non-negotiables here. You may not be a big fan of Steve Gould but are you willing to draw a line in the sand by saying that Darwinism (or evolution) will be falsified if Gould turns out to be correct?

Joe Felsenstein said...

Even if one folds mutation, recombination, inter-population gene flow, inter-specific introgression, etc into variation Dembski has still omitted genetic drift. While natural selection is probably the predominant proximate cause of adaptation it's rather harder to say that the roles of cladogenesis and genetic drift in providing a substrate for exaptation can be neglected.

Disagree. No matter what genetic drift and other evolutionary forces are doing, they themselves cannot explain why genomes are much better-adapted than a random string of nucleotides. For that you need natural selection. I am not saying natural selection is all-powerful, and I am not saying at what level the selection occurs. But there simply is no other evolutionary force that biases the genome toward adaptation.

That is also why the "mutationists" of the early 20th century were very wrong: they could not show why mutation would lead to better adaptation, without natural selection around to do that. It is also why present-day mutationists (James Shapiro?) are equally wrong. However much the other forces do to the genome, the presence of nonrandomly high amounts of adaptation can only be explained if you have natural selection around to do that.

konrad said...

Agreed - all animals, even members of the same species, are qualitatively different. This is mainly but not exclusively due to genetic differences, but even identical twins are qualitatively different. So what on earth is Dembski talking about here?

alias Ernest Major said...

No and no.

Picking the right words to avoid ambiguities and misunderstanding is not always easy, but I don't see where your misinterpretation comes from.

In this context regularities means that there is a (at least statistically) consistent relationship between cause and effect. For example, similar meteor strikes have similar effects. I'll also point out that I referred to processes, not events - I was making a statement on what processes can be considered by science, not what events can be considered by science.

The qualification about statistical regularities is to avoid unintentionally excluding processes affected by quantum indeterminism and deterministic chaos from the scope of science.

MRR said...

I think that "non-negotiables" can be understood in two different ways here:

1- If you challenge these points without evidence, you are placing yourself in opposition to science. The list could be improved, but it's not too bad.

2- Evolutionary biology cannot refute these points without denying its own foundations. There are no such "non negotiables" in science, as several commentators have pointed out.

I think that meaning 1 is useful to the discussion of compatability between science and religion.

Claudiu Bandea said...

I am not saying natural selection is all-powerful...

I don’t know Joe Felsenstein, but based on his two paragraphs above, definitely, he understands evolution. I’m highly impressed! However, I would say that natural selection is all-powerful. For why I’m saying that, stay tuned!

Larry Moran said...

Joe Felsenstein (Happy Birthday!) says,

I am not saying natural selection is all-powerful, and I am not saying at what level the selection occurs. But there simply is no other evolutionary force that biases the genome toward adaptation.

I gave you a quotation from Michael Lynch (above) who talks about the nonadaptive evolution of complexity. The idea is that there are features of complex genomes that count as adaptations by any reasonable criteria but these features arose as the result of random genetic drift.

Here's how Eugene Koonin describes it in The Logic of Chance (page 242).

The evolution of advanced adaptations in small populations with weak selection might appear paradoxical, and perhaps for good reason: Evolving such complex features seems to require efficient positive selection that is possible only in populations with a large Ne. This is admittedly a difficult problem. The solution, however counterintuitive, seems to require "weak anthropic reasoning": Species in which these complex features have not been fixed, primarily via random genetic drift and constructive neutral evolution ..., simply had no chance to survive.

I can understand why you may disagree. What I don;t understand is why you are so certain that Koonin and Lynch have to be wrong. Are you saying that if they are right then one of the "non-negotiables" of evolutionary theory has been falsified? Would that mean that the theory of natural selection (or Darwinism?) has been falsified because it requires that natural selection be the only force capable of producing adaptation?

Larry Moran said...

Claudiu Bandea says,

I don’t know Joe Felsenstein, ...

Check out his Wikipedia reference [Joseph Felsenstein]. Among other things, you'll learn that he was a "red diaper baby" and that he was named after another famous "Joseph."

Claudiu Bandea said...

Check out his Wikipedia reference [Joseph Felsenstein]

Thanks for the reference. Anyone babbling in evolution knows about that Joe Felsenstein, and I did too, but I did not recognize his picture, and I did not expect to have him on your blog Larry (not pounding intended here, just surprise!).

Among other things, you'll learn that he was a "red diaper baby" and that he was named after another famous "Joseph."

Well, I think that’s not his (in alphabetical order) fault or feat, so I would leave it at that.

I gave you a quotation from Michael Lynch (above) who talks about the nonadaptive evolution of complexity

Species in which these complex features have not been fixed, primarily via random genetic drift and constructive neutral evolution ...

The two concepts, nonadaptive evolution of complexity and constructive neutral evolution are just silly and oxymoronic concepts, respectively. Again, stay tuned for an alternative take on natural selection and evolution, a take intended by Darwin but misinterpreted by many of his followers (I think, not including Dr. Felsenstein) and non-followers.

Larry Moran said...

... I did not expect to have him on your blog Larry (not pounding intended here, just surprise!).

Do you realize how many people you've just insulted (besides me)?

Larry Moran said...

The two concepts, nonadaptive evolution of complexity and constructive neutral evolution are just silly and oxymoronic concepts, respectively. Again, stay tuned for an alternative take on natural selection and evolution...

Wow! You set yourself a pretty high bar by dismissing some very smart scientists as "silly."

I doubt that you can live up to those expectations.

Joe Felsenstein said...

Eugene Koonin quoted by Larry: Species in which these complex features have not been fixed, primarily via random genetic drift and constructive neutral evolution ..., simply had no chance to survive.

Larry's comment: I can understand why you may disagree. What I don;t understand is why you are so certain that Koonin and Lynch have to be wrong. Are you saying that if they are right then one of the "non-negotiables" of evolutionary theory has been falsified? Would that mean that the theory of natural selection (or Darwinism?) has been falsified because it requires that natural selection be the only force capable of producing adaptation?

That example is one level of natural selection ... it's species selection. Species that don't achieve that adaptation are more likely to go extinct.

What I am saying is that if living organisms are better at living than random chunks of primordial slime (or than whatever would be coded for by totally random DNA sequence), then the only explanation for that is some form, some level of natural selection. Other evolutionary forces come in and have all sorts of effects, but by themselves they cannot make fish that swim and birds that fly. Not without some natural selection.

Mutationists got this wrong (way back before the Modern Synthesis) and present day mutationists get it wrong still. Larry, are you a mutationist?

And by the way, thanks for the birthday greeting!

Claudiu Bandea said...

Do you realize how many people you've just insulted (besides me)?

I don’t know if your response is intended as a payback 'pun,' or you are serious (I’m not very advanced in the art of blogging to be able to differentiate). But just to clarify, my remark was intended that way, as a 'pun'!

Wow! You set yourself a pretty high bar by dismissing some very smart scientists as "silly."

I think that these scientists are not only smart, but they are some of the most profound and productive thinkers in modern biology. Period! However, I also think that on this particualr issue they happen to be wrong, or 'silly'. But, maybe I’m wrong and silly!

john harshman said...

Species in which these complex features have not been fixed, primarily via random genetic drift and constructive neutral evolution ..., simply had no chance to survive.

That's why Joe said he wasn't specifying the level of selection. What's described here is just species selection, a form of group selection. Still selection, though.

Barbara said...

Yes, I'd make some modifications. The most important would be the label. It may suggest these are assumptions scientists started out with, but at their most certain, these "non-negotiables" are conclusions from careful observations and testing. They might not have been true (and aren't entirely true as stated). But to the extent they are true (as well as we can tell) they're "non-negotiable" until good evidence that they're false shows up.

As to specifics:

D1: All organisms really are related to common ancestors by descent with modification (conclusion, not assumption), and on earth we probably do trace back to a single ancestor, but in theory there might have been two or more origins.

D2: (It's not actually false, but ignores the crucially creative role of genetic drift.)

D3: Again, this one manages to be not really false, but to ignore important things, and uses an ambiguous, emotionally loaded "fundamental." We humans are continuous with other animals. Our structures, physiology, and the basic functioning of our brains are fundamentally the same as corresponding traits of other animals. And yet out of that fundamental sameness we have also produced complex thoughts, technology, and societies that no other organism can achieve. This difference can be considered a difference in kind, although it results from extreme differences in degree. Many people consider this difference "fundamental" to being human and so will find this "non-negotiable" example offensive.

D4: As scientists we couldn't assume that the physical world operates by "unbroken natural law," but three centuries or more of careful observation strongly suggests that that's true. Therefore, at this point we have to assume that non-natural explanations are irrelevant, unless they're supported by extraordinarily strong evidence.

Larry Moran said...

Evolutionary biology cannot refute these points without denying its own foundations. There are no such "non negotiables" in science, as several commentators have pointed out.

Most of us think that any theory can be falsified and discarded by the discovery of new information. What does that mean? It means that any useful theory must have "non-negotiables" whose falsification would cause us to look for a better theory.

The question before us is, what are those core tenets of modern evolutionary theory?

Your statement doesn't make sense. You seem to agree that Dembski's list of non-negotiables would, if falsified, deny the very foundations of evolutionary theory. That's correct.

Then you turn around and state categorically that there's no such thing as "non-negotiatables" in science. Can you justify those apparently contradictory statements?

Larry Moran said...

Hmmm ... Stephen Jay Gould wrote a very large book advocating species sorting (not species "selection"). He explains why it's not the same as natural selection.

Are you saying that natural selection—generally assumed to be a population level phenomenon—alos covers species sorting where selection isn't even involved?

Do you know of any textbook on evolutionary biology that adopts this position or are you simply advocating that they do so as quickly as possible?

Larry Moran said...

... but in theory there might have been two or more origins.

In other words, D1 is not a "non-negotiable" of modern evolutionary theory in spite of what Dembski says?

Therefore, at this point we have to assume that non-natural explanations are irrelevant, unless they're supported by extraordinarily strong evidence.

I agree that's a reasonable, but tentative, conclusion of science. It is not a restriction and it is not a "non-negotiable" of science. However, the more I think about it, the more I can see that it might actually be a "non-negotiable" of modern evolutionary theory.

If we were to discover supernatural intervention in evolution then, at the very least, evolutionary theory would have to be modified to include a mechanism of change in allele frequencies that wasn't natural selection or random genetic drift. That's a very drastic change.

We would be justified in saying that our current version of evolutionary theory has been refuted.

Larry Moran said...

Joe Felsenstein says,

That example is one level of natural selection ... it's species selection. Species that don't achieve that adaptation are more likely to go extinct.

I don't think species sorting (or its subset, species selection) is the same thing as natural selection. But I see where you are coming from. If you broaden the mechanism of "natural selection" to include anything that can eventually give rise to an adaptation then your claim is true, by definition.

Do you know of any other evolutionary biologists who make such a claim? Is it in any of the textbooks on evolutionary biology?

Mutationists got this wrong (way back before the Modern Synthesis) and present day mutationists get it wrong still. Larry, are you a mutationist?

I'm very sympathetic to modern mutationism so, yes, I would define myself as a mutationist but perhaps not as dedicated to that position as some of the leading evolutionary biologists.

For those of you who don't know much about the modern theory of mutationism, I recommend that you read the excellent series of postings by Arlin Stoltzfus: The Mutationism Myth. You won't be qualified to discuss modern mutationism unless you understand it and you won't find a better summary than those postings.

I think that what Joe is saying is that if mutationism is true then modern evolutionary theory has been refuted. However, if Koonin and Lynch are correct then modern evolutionary theory won't be falsified because those evolutionary theorists don't understand that their proposals are perfectly consistent with natural selection. (In spite of the fact that they say otherwise.)

John Harshman said...

Well, as Gould says, it's not quite clear that Darwin meant only individual selection by that term. While it's true we mean it that way now, I think Joe's specific disclaimer shows what he meant. What do you think that disclaimer meant?

As for the difference between species selection and species sorting, I personally don't see it. There's long been a claim that species selection can only act on emergent properties, but I see no reason for such a demand. I would claim that any fixed difference between species, even if it's just the aggregate of individual characters, can if it causes differential extinction or speciation be considered a subject of species selection. I agree that this isn't the same as individual selection ("natural selection"); obviously, as its "individuals" are species.

If we're just arguing about what term to apply to selection whose level -- individual, deme, species, etc. -- hasn't been specified, I don't see the argument as at all interesting. Gould called it multi-level selection. I don't care what we call it, as long as we understand what we're talking about.

Joe Felsenstein said...

I just don't see some big qualitative difference between species selection and group selection. If group selection is a kind of natural selection, isn't species selection that too?

We do talk about "levels of selection". Why, when we get to species, are we supposed to consider it "sorting" instead of selection?

Larry: If you broaden the mechanism of "natural selection" to include anything that can eventually give rise to an adaptation then your claim is true, by definition.

My use of "natural selection" to cover all these does not render my argument circular. I am lumping under natural selection all processes that are biased in the direction of adaptation. An adaptation that happened accidentally by (say) genetic drift would not be included, so my argument is uncircular.

Anonymous said...
This comment has been removed by the author.
Anonymous said...

Why the fuck would anybody even propose "non-negotiables"? Science is not about to start proposing dogmas. What an ass-hole would somebody have to be in order to propose that something like "Darwinism" would/should be dogmatic?

For one, if by "Darwinism" is meat "Darwin's propositions", I doubt that Darwin would have proposed anything as a "non-negotiable." For another, if by "Darwinism" is meant "evolution," I doubt that a scientist studying or reflecting on how evolution happens would/should start with "non-negotiables." I see no reason to play Dembski's game. I see no reason why we should start allowing ass-holes such as Dembski to define our terms and our way of thinking. He can go get fucked. (Yes, the latter is a non-negotiable.)

Claudiu Bandea said...

Back to Darwin! In my interpretation, he said:

Dw1. All organisms on Earth are evolutionary related

Dw2. Natural selection drives evolution

In Dw1, all organisms or species, if we group them that way, meant 'all', which includes humans. And, because organisms/species are different from each other, obviously, they are also different from their ancestors, and their descendants will be, more or less, different from them (i.e. descent with modification).

In Dw2, the concept of 'natural selection' was based on his observations that the selective forces leading to evolution, which implies adaptation, were ‘natural,’ not ‘divine’ as defined by religious organizations at the time he lived

Larry Moran said...

John Harshman says,

As for the difference between species selection and species sorting, I personally don't see it.

Let's say you observe in the fossil record that some of the closely related species in a clade go extinct. You don't know whether this was by chance or whether it was due to selection. In that case the proper term is "species sorting."

Why is that so difficult to understand?

I don't care what we call it, as long as we understand what we're talking about.

Hmmm .... I think definitions are important. You can't just make up your own definition like Humpty Dumpty. I agree that once you explain that your definition of natural selection is different than others then we can understand each other. However, it would be much better if we all agreed on a common definition of "natural selection," don't you think?

Larry Moran said...

Joe Felsenstein asks,

If group selection is a kind of natural selection, isn't species selection that too?

I suppose, but does everyone agree that group selection is just another kind of "natural selection"?

Why, when we get to species, are we supposed to consider it "sorting" instead of selection?

See my response to John Harshman. We use the term "species sorting" because we don't know whether selection is involved. It could just be accident.

If the survival of one species and not another turned out to be entirely due to chance events, would you call that an example of random genetic drift?

An adaptation that happened accidentally by (say) genetic drift would not be included, so my argument is uncircular.

Now I'm really confused. In your earlier comment on May 10th you said ...

Disagree. No matter what genetic drift and other evolutionary forces are doing, they themselves cannot explain why genomes are much better-adapted than a random string of nucleotides. For that you need natural selection. I am not saying natural selection is all-powerful, and I am not saying at what level the selection occurs. But there simply is no other evolutionary force that biases the genome toward adaptation.

To me that says that the only force capable of adaptation is natural selection. You seemed to disagree with Lynch, Koonin, and others who postulated that drift could lead to adaptation.

Have you changed your mind or is there some nuance that I'm not getting?

Joe Felsenstein said...

Yes, there are nuances you aren't getting.

1. Do you allow kin selection as a form of natural selection? I do. And the equations for the success of group selection (in some generic cases) are the same as Hamilton's equation for kin selection. I would find it hard to draw a line between kin and group selection and say that one "is" natural selection and the other isn't.

2. Random change due to random speciations and extinctions needs a name. I would not call it genetic drift. Maybe "species drift".

3. Drift (of either type) can lead to some adaptation. But long-continued adaptation could not result from drift. That's why I qualified my statement by saying "much better-adapted". Fish gotta swim and birds gotta fly, as Oscar Hammerstein wrote. And if they can do so a lot better than a random pile of primordial ooze, that must be due to one or another form of selection. And it is not doubtable that they do swim and fly a lot better than that.

john harshman said...

Let's say you observe in the fossil record that some of the closely related species in a clade go extinct. You don't know whether this was by chance or whether it was due to selection. In that case the proper term is "species sorting."

Why is that so difficult to understand?


Because it's a weird thing to say. In exactly the same way, you don't know whether the fixation of any particular allele is due to chance or selection. All you can do is consider the plausibility or, if there's some kind of prior prediction and model, an estimate of probability. Some outcomes are unlikely through drift. The closest I can come to agreeing with you here is that "population" and "sample" sizes are generally much smaller when species are the units than when individuals are the units, and so harder to test models on. But that had nothing to do with the reason Vrba originally proposed "species sorting", which under her scheme wasn't ambiguous but was most certainly not species selection.

john harshman said...

By the way, this quote presupposes species selection, i.e. that extinction is due to presence or absence of particular characteristics, not chance.

Species in which these complex features have not been fixed, primarily via random genetic drift and constructive neutral evolution ..., simply had no chance to survive.

Larry Moran said...

John Harshman says,

By the way, this quote presupposes species selection, i.e. that extinction is due to presence or absence of particular characteristics, not chance.

Koonin is referring to evolvability and that's not the same as species sorting.

Species sorting occurs when you consider a clade (related species) as a "population" and individual species as an "organism." The differential survival of species within such a clade can be due to species selection or species drift.

Koonin's examples have more to do with evolution at a larger scale. For example, the evolution of a group of animals with segmented body parts under the developmental control of HOX genes may have enabled that class to survive and speciate—over hundreds of millions of years—better than the animals who didn't have such a pattern of development. The important point is that the evolution of such a complex developmental pathway may not have had any immediate selective advantage for the first species who evolved it.

The evolution of placental mammals and their preferential survival over marsupials in most of the world is another possible example.

Larry Moran said...

John Harshman says,

But that had nothing to do with the reason Vrba originally proposed "species sorting", which under her scheme wasn't ambiguous but was most certainly not species selection.

Here's how Gold describes species sorting in his 2002 book (page 659).

... Vrba and I developed a terminology to resolve a common confusion in evolutionary theory between the simple, and purely descriptive, observation of differential reproductive success—which we named "sorting"—and the causal claim—always and properly called "selection"—that the observed success arises from interaction between properties of the the relevant evolutionary individual and its environment (see Vrba and Gould, 1986). Evolutionary biology needed this distinction because students of the field have often—with misplaced confidence in selection's ubiquity and exclusivity—made a case for selection based on nothing more than an observation of reproductive success (sorting), without any attempt to elucidate the cause of such sorting.

He then goes on to explain that drift is the other important option that is often overlooked. Thus species sorting and species drift are two possible mechanisms that can explain species sorting.

John, I don't see why you are resisting this fairly obvious terminological distinction. Am I missing something?

🐰 said...

ET already needs a mechanism other than NS, drift and ~random mutations to account for the changes in allele frequencies caused by human genetic engineering. Presumably "supernatural" genetic engineering could be handled the same way.

Joe Felsenstein said...

Of course we should not jump to the conclusion that selection is responsible whenever we see a net change at some level (species, population, individual). But the cases that John Harshman and I are pointing to are where properties that depend on the genotype of the organism do lead to the outcome. My reading of the Gould and Vrba passage is just that thay want a name for cases where a change might just be a random outcome of random speciation and extinction.

In the cases where the outcome is not the result of that "species drift", I think it is sensible to call this natural selection. But if Larry wants to call species selection something else, fine. The point of Dembski's (D2) and of my own argument is that some nonrandom process that is biased in favor of reproduction of more adapted individuals must be involved if we see substantial improvement of adaptation over the long term.

And yes, there has been substantial improvement of adaptation when you consider net change since the organic soup.

Quibbling about whether you should call it by several different terms (depending on the level) does not make the point go away.

Arlin said...

I'm so sorry that I missed this discussion. I'm disturbed by the attitude of several respondents who can't stomach the idea that "Darwinism" has non-negotiable parts, or includes doctrines. How could a scientific theory be falsifiable if it all its parts were negotiable? If we allowed the meaning of "Lamarckism" to be negotiated far enough, we would all be Lamarckians.

Darwin's _natura non facit saltum_ is clearly a doctrine. It can hardly be argued that Darwin proved this, or reached it as an inductive generalization. It simply was part of Darwin's DNA to believe that great things happen in small doses. Jacques Barzun (in Darwin, Marx, Wagner) argues that this belief was very common in Darwin's generation. This is part of what makes his theory what it is. To deny this doctrrine is to deny what the man actually intended to convey in his theory of evolution.

I think Dembski's list is quite good for such a succinct and high-level description. It does not go far enough to describe Darwin's scientific theory, however. Darwin made several very clear statements about his theory. One of them, following from his _natura non facit saltum_ doctrine, was that it would "break down" if it could be shown that some feature could not have arisen by numerous successive slight modifications.