More Recent Comments

Friday, December 13, 2013

What do they mean when they say they want to extend the Modern Synthesis?

As far as I'm concerned, the "Modern Synthesis" has been replaced by modern evolutionary theory that incorporates Nearly-Neutral Theory and random genetic drift as an important mechanism of evolution [see Is the "Modern Synthesis" effectively dead? ]. This extension, and replacement, of the 1940s version of evolutionary theory took place in mainly in the 1970s.

If I'm correct, then why all the fuss in the 21st century about extending the Modern Synthesis?

I think there are two things going on here. First, there are a bunch of biologists who want to incorporate their favorite fad into modern evolutionary theory. They think that their ideas are so revolutionary that this requires an extensive revision of evolutionary theory. Second, those biologists seem to have been asleep during the 1970s when the Modern Synthesis died so they are fighting a strawman.

There's an entire book devoted to these fads. The editors are Massimo Pigliucci and Gerd B. Müller. Here's what they write in chapter 1.
More than a century has passed since the integration of several strands of evolutionary thought into what came to be called Modern Synthesis (MS), the conceptual framework that has defined evolutionary theory since the 1940s. Despite significant advances since then in all methodological and disciplinary domains of biology, including molecular genetics, developmental biology, and the "-omics" fields, the Modern Synthesis framework has remained surprisingly unchanged. Although it is still regarded as the standard theoretical paradigm of evolutionary biology, for several years now dissenters from diverse fields of biology have been questioning aspects of the Modern Synthesis, and pivotal novel concepts have been elaborated that extend beyond its original scope. As a result, calls for an expansion of the Modern Synthesis framework have intensified, prompting further scientific debate.
It seems as though Pigliucci and Müller are unaware of the revisions that took place in the 1970s or else they think these were insignificant allowing the 1940s theory to remain "substantially unchanged."

Either way, they look awfully foolish to me.

Most of you know my views on the Central Dogma of Molecular Biology [Basic Concepts: The Central Dogma of Molecular Biology]. I have very little patience with those scientists who criticize the Central Dogma without ever bothering to find out what it means and without ever reading the papers that Crick wrote. I think that such sloppy scholarship is a pretty good indication of the quality of scholarship those scientists apply to other topics.

Let's see how Pigliucci and Müller handle the "Central Dogma Test" a few pages later.
As we will see in the rest of this volume, several of these tenets [of the Modern Synthesis] are being challenged as either a inaccurate or incomplete. It is important however, to understand the kind of challenge being posted here, in order to avoid wasting time on unproductive discussions that miss the point of an extended evolutionary synthesis. Perhaps a parallel with another branch of biology will be helpful. After Watson and Crick discovered the double-helix structure of DNA, and the molecular revolution got started in earnest, one of the first principles to emerge from the new discipline was the unfortunately named "central dogma" of molecular biology. The dogma (a word that arguably should never be used in science) stated that the flow of information in biological systems is always one-way, from DNA to RNA proteins. Later on, however, it was discovered that the DNA > RNA flow can be reversed by the appropriately named process of reverse transcription, which takes place in a variety of organisms, including some viruses and eukaryotes (through retrotransposons). Moreover, we now know that some viruses replicate their RNA directly by means of RNA-dependent RNA polymerases, enzymes also found in eukaryotes, where they mediate RNA silencing. Prions have shown us how some proteins can catalyze conformational changes in similar proteins, a phenomenon that is not a case of replication, but certainly qualifies as information transfer. Finally, we also have examples of direct DNA translation to proteins in cell-free experimental systems in the presence of ribosomes but not of mRNA. All of these molecular processes clearly demolish the alleged central dogma, [my emphasis - LAM] and yet do not call for the rejection of any of the empirical discoveries or conceptual advances made in molecular biology since the 1950s. Similarly, we argue, individual tenets of the Modern Synthesis can be modified, or even rejected, without generating a fundamental crisis in the structure of evolutionary theory&mdas;just as the Modern Synthesis itself improved upon but did not cause the rejection of either Darwinism or neo-Darwinism.
I'm more than willing to accept their parallel. I give just as much credence to their attack on the "Modern Synthesis" as I do to their attack on the Central Dogma of Molecular Biology.

(That's "no credence at all" for those of you who don't appreciate the irony.)


DGA said...

"'When I use a word,' Humpty Dumpty said in rather a scornful tone, 'it means just what I choose it to mean — neither more nor less." Call this the HD Principle, it seems to be what you have a problem with - people seem to hear a phrase, incorrectly, incompletely, inattentively and then redefine it to suit whatever they want it to mean or think it must mean based on their faulty recollection.

DGA said...

I mean that with regards to the misstating of the Central Dogma.

Alex SL said...

Maybe this is going to sound silly but I do not quite understand this concern about the various syntheses. On the one side there are people hoping to be the next paradigm-destroyer or to become famous by making big claims, and the people obsessing about whether somebody made a minor mistake in spelling out the official definition of this or that, and those quarreling whether this or that is the right metaphor for evolution.

And on the other side we have the question how evolution actually proceeds. This - and perhaps only this - is what we should be concerned with. I have never in my life had the need to talk about the Modern Synthesis (tm) or Central Dogma (c) with students or collaboration partners, nor will I have any use for an Extended Synthesis (patent pending), as long as I have a decent understanding of population genetics, selection, drift, lineage sorting, basic biochemistry, biogeography and suchlike.

Georgi Marinov said...

I guess we need the labels because what lies behind many of them is lack of decent understanding of the things you listed.

Joe Felsenstein said...

I agree that this is more about labels than about what processes are actually responsible for evolution. We would not need the labels at all except that people keep claiming that discovery of one phenomenon or another invalidates the Modern Synthesis (or, in worse terminology "neo-Darwinism"). This has been going on for years now; the closest to an actual invalidation was Punctuated Equilibrium in the 1970s-1980s. But it is iffy whether that would invalidate the Modern Synthesis, and besides the centrality of PE is not widely agreed on.

The downside of agreeing too readily on the death of the Modern Synthesis is that it leaves the public confused, thinking that the processes they heard of in school now are no longer thought to be important (like natural selection).

I can perhaps best make the point by imagining someone claiming, in about 1860, that Newton's Laws of Motion were no longer a live item. After all, the physics of that time was actively concerned with phenomena that Newton never imagined -- thermodynamics and electricity, for example. So the Newtonian view was dead then, right?

Or imagine someone claiming in about 1940 that Mendelian Genetics was dead. After all, we knew then of phenomena Mendel never imagined -- chromosomes, linkage, multiple alleles, sex chromosomes, even mutation itself. So bye-bye to the Mendelian view, right?

So I think we should be resistant to relabeling, when the package of processes has remained valid, though individual forces have had their examples expanded. With all our greater understanding and all our powerful molecular biology, the chief evolutionary forces today are still the ones that the founders of the Modern Synthesis saw in about 1930 -- genetic drift, mutation, migration, and yes, natural selection.

And their view then was substantially different from what the picture had been a few decades before 1930, when mysterious orthogenetic forces, mysterious saltational jumps, and Lamarckian views were widely credited, and natural selection was in the doldrums. Our view today is closer to 1930, and still very far from what the picture was in 1900. That reflects a change to a new view back then, one which is still valid.

Unknown said...

Having read this article by Dr. Moran (and two others it points to) I have two questions, or more accurately two reformulations/extrapolations of what was said that I want to be sure are correct.

1) In summary Larry's position is: The original 30's modern synthesis (MS) was built around defining evolution in terms of allele frequencies which comfortably permits (near neutral) drift. In a later "hardening" (incarnated by Mayr ??) the MS was reinterpreted to exclude anything but selective "forces" as responsible for change. So the original MS is just fine, although some of the details have since been filled in. ??

2) In recent discussions with creationists I have come close to needing to discuss drift. I have also recently noticed this:

From the diagram I note that the degree of difference in the DNA of morphologically distinct dolphins vs. cows is about the same as that between the (apparently) more similar mouse vs. rat. This appears to me to be a nice visual presentation of the effects of drift, since selection driving greater morphological change has been achieved with the same amount of genetic change.

Is this example an informative illustration of the effects of genetic drift?

John Harshman said...

I'm not sure I understand what you think is drifting here. The bulk of divergence is due to drift, and a small proportion of divergence has anything to do with morphology, though that part is mostly under selection. What your example shows is that most of the genome isn't under selection. Is that what you meant?

Unknown said...
This comment has been removed by the author.
Larry Moran said...


My understanding of the history of population genetics is that by the 1940s very few believed that random genetic drift was important. It might play a minor role in moving species off an adaptive peak but, even then, it was not widely accepted.

I think there's an enormous difference between the way we view random genetic drift today and how it was perceived in the 1940s. You'll note that it gets no attention in Huxley's book.

We've both been to meetings of evolutionary biologists. You know full well that a substantial percentage of them have an extremely adaptationist perspective. They never even consider the possibility of drift in their publications.

I just took a MOOC on evolution taught by evolutionary biologists at the Natural History Museum in New York. The course was intended for teachers who have to teach education. Evolution was not defined and there was no mention of random genetic drift as an important mechanism When I raised this issue on the forums, it was clear that the vast majority of students (teachers) had no idea what I was talking about.

I think you are underestimating the importance of this problem and I think you are incorrect when you claim that the current version of the Modern Synthesis includes drift and Nearly-Neural Theory. That's not how most people see evolutionary theory.

If you haven't read the book edited by Pigliucci and Müller then I think you should. Let me know if you think the majority of authors really understand evolution the way you do.

Arlin said...

I could write a book about this, and someday I will. Every scientist with a PhD knows that it takes years of study to be an expert-- you not only have to ingest hundreds of books and papers, you have to digest this information, sifting it and evaluating it, and converting it into knowledge and judgment. Yet evolutionists all think they are automatic experts on the topic of "the Modern Synthesis" or "the history of evolutionary thought" without doing original research on it, or reading hundreds of secondary sources.

Scientists' views of their history are based largely on b***s*** written by other scientists in the Introduction and Conclusion sections of their papers, and books written by old scientists who claim to have authority because they were there. Ernst Mayr, who has been far too influential, was a Lamarckian systematist pondering birds in a dusty museum during the time that the 20th century genetical view of evolution was being built by others-- people conducting genetical experiments, analyzing data using statistics, and developing mathematical models that Mayr understood only secondarily, through the interpretations of others. He wasn't there. Mayr's version of history was basically to tell unflattering stories about how intellectual opponents were misled by prejudices, due to "essentialism" or "typology" etc. Historians have thoroughly discredited Mayr's main ideas. Although Gould actually did read extensively about the topic of history, his magnum opus attributes some kind of psychological problem to every single scientist who comes out against Darwin. Shame on Gould.

To know what the Modern Synthesis means, you actually have to do a research project that takes years. That sucks. In an ideal world, what the "Modern Synthesis" means would be clearer. But it isn't. I could explain it in a few sentences, but you wouldn't believe me because the explanation has to come with a huge justification.

The reason that so much justification is needed is that there are droves of scientists who-- for reasons that I can't fathom-- imagine they are doing the right thing by defending the idea that the Modern Synthesis covers whatever we think today, without bothering to consider that proposition skeptically, as a hypothesis. In fact, the mainstream of evolutionary thinking today includes views that were considered "non-Darwinian" and "anti-Darwinian" a century ago, such as the view that evolution depends frequently on new mutations, or the view that selection is not creative but merely sorts out pre-existing genetic varieties.

To understand history, one must begin by tossing out a false conception of change as a set of continuous shifts along pre-existing dimensions (i.e., Darwinists think history works in the same way they think evolution works). The history of evolutionary thought cannot be described by moving a variety of sliders. For instance, today most of us think that an allele fixation (by selection or drift) is the unit event in evolution, and we tend to think of evolution as a Markov chain of such events. It makes sense in this view to tally up fixations by drift vs. fixations by selection. This is a familiar idea to us. A generation of evolutionists were raised with the idea that the most important question in evolution is what fraction of changes are due to drift vs. selection. However, 70 years ago this question would not have made sense because they just did not think about evolution in this way. The Modern Synthesis was very much a reaction against the idea that evolution can be understood as 2-step origin-fixation process. That was "mutationism". I see no evidence that biologists 70 years ago had a clear concept of fixations by drift as an evolutionary process. So, with apologies to Larry, we cannot understand the history of evolutionary thinking by asking questions like this.

Unknown said...

John Harshman;

In response to:
"I'm not sure I understand what you think is drifting here."

After some reflection, I suspect there are many ways I could unwittingly go astray. As explicit as I can make it, here is my understanding:

For present purposes the genome can be divided in two:
1) The functional genome: anything that codes or regulates and thus has biological/morphological effects.

2) Nonfunctional genome: everything else.

Although I read all the words in the entire paper, I didn't understand at least 2/3rds of it. I don't know what "based on the substitution rates in the MultiZ alignments" means, but the notes for figure 1 appear to indicate that the tree might have been built from [an average of] 4.2% of each genome, which would be the functional genome.

Whether or not the tree was built from all of the genome that could be decoded, or just the functional part, genomic elements that have no effect (like introns and substitutions that don't affect the function of proteins etc) will still account for most of the length of the tree arms separating pairs like mice/rats, dolphins/cows etc.

Thus considered as a characteristic, genomic encoding (expressed or not), is by far the largest element not under selection and thus subject to drift. So "yes" in answer to your second question.

I wonder if "adaptationist" evolutionary biologists think that genomic encoding is not a characteristic?

I must admit that I become seriously confused when considering the possibility of drift in expressed morphologic characteristics. I find myself going around in circles between definitions and real world events. Given my objectives, I don't think it matters if I sort this issue out, and I seriously doubt that I have the mental ability to do so.

John Harshman said...

Multi-Z is the program they used to make alignments. The paper and the supplementary information are not clear, but it appears that the tree topology was fixed in advance based on an unspecified source, while the branch lengths are based on all 4-fold degenerate sites in all candidate protein-coding exons. That is, the branch lengths are supposed to be mostly from neutrally-evolving sites, though not the ones you mention.

I'm not sure what you mean by "decoded", but you probably mean "aligned", that is with homologous sites matched up to each other across species. I'm also not sure what "genomic encoding" means to you, and so neither of the sentences containing that phrase are decipherable for me.

There can easily be drift in a morphological character if none of the alternative states of that character is significantly advantageous compared to the others. ("Significantly" depends on population size.) Finding real world examples, and distinguishing them from examples of selection, would require considerable experiment and/or observation of populations.

Unknown said...

John Harshman;

Concerning your statement:
"while the branch lengths are based on all 4-fold degenerate sites in all candidate protein-coding exons. That is, the branch lengths are supposed to be mostly from neutrally-evolving sites, though not the ones you mention."

You appear to be telling me that within the 1-2% of the genome composed of protein-coding exons (presumably for proteins that are found in most or all mammals) there are regions that can change in content (but rarely in length) without much affecting the function of the resulting protein. It was differences in these regions that was used to establish tree branch lengths between different pairs/groups of species.

Concerning your statement:
"it appears that the tree topology was fixed in advance based on an unspecified source"

Probably unimportant, but is it possible to derive a different topology from the data used? If not why does a question concerning the source of the topology even arise?

Please forget about the phrases that used the words "decoded" and "encoded" however in both cases I intended something like "could be read and positioned on a chromosome, whether or not they were functional"

Thank you very much for your explanation, without which I might have wandered into all sorts of foolishness.

John Harshman said...

You appear to be telling me that within the 1-2% of the genome composed of protein-coding exons (presumably for proteins that are found in most or all mammals) there are regions that can change in content (but rarely in length) without much affecting the function of the resulting protein. It was differences in these regions that was used to establish tree branch lengths between different pairs/groups of species.

As usual, I have some problems with your terminology, but yes, any base at 4-fold degenerate site gives you the same amino acid, so there can be no selection on them, at least selection that works through the part of the phenotype that consists of proteins. DNA sequences too are part of the phenotype, and there may be a bit of selection there. But to a first approximation, there isn't.

Probably unimportant, but is it possible to derive a different topology from the data used?

We don't know. They didn't do the analysis to find out, as far as I can tell.

"Could be read and positioned on a chromosome".

Hmmm...I think by this you refer to what we call sequencing and assembly.

Mong H Tan, PhD said...

RE: Can neo-Darwinists or “pseudogenetic evolutionists” self-reflect, retract, or correct their very flawed “disconnect” theory of Modern Synthesis (MS) since the 1930s-40s!?

My short answer is No -- as I briefly analysed at the Sandwalk discussions herein before!?

Furthermore -- unlike Arlin Stoltzfus -- most “evolutionist-trained biologists” or “pseudogenetic evolutionists” are incapable of or at best unwilling to reflect or retract or correct their predominantly academically-received “evolutionary dogma and rhetoric” as one that is MS -- the MS that has had been so popularised, sensationalized, and fixated by the 2 world’s most renowned neo-Darwinist evolutionists and biologists, EO Wilson and Richard Dawkins, since the mid-1970s!? -- [Please see their respectively published neo-Darwinist evolutionist books Sociobiology: The New Synthesis (1975) and The Selfish Gene (1976).]

As such, the argument put forward by their subsequent or parasitic neo-Darwinists Pigliucci and Müller that “individual tenets of the Modern Synthesis can be modified, or even rejected, without generating a fundamental crisis in the structure of [the neo-Darwinian (edits mine; MHT)] evolutionary theory -- just as the Modern Synthesis itself improved upon [by covertly subsuming Mendelism while without acknowledging it] but did not cause the rejection of either Darwinism or neo-Darwinism” clearly shows that they both have had accepted the MS “disconnect theory” unequivocally and uncritically; and therefore they are now (2010) and here only to improve or extend it -- just as Dawkins has had “extended” his “selfish gene” anthropomorphic rhetoric (not science nor genetics) to his subsequently MS-inspired “extended phenotype” synthetic theory (1982)!? Or, the Wilson’s also MS-inspired “myrmecology extended to both anthropology and sociology thesis” as his then newly synthesized (also disconnectedly) sociobiology theory since 1975!?

Whereas the original MS “disconnect” evolutionary theory has had been delfly debunked by Stoltzfus elsewhere and above before!?

Best wishes, Mong 1/9/14usct2:50p; practical science-philosophy critic; author "Decoding Scientism" and "Consciousness & the Subconscious" (works in progress since July 2007), Gods, Genes, Conscience (iUniverse; 2006) and Gods, Genes, Conscience: Global Dialogues Now (blogging avidly since 2006).