More Recent Comments

Tuesday, February 02, 2016

What is "structuralism"?

The Intelligent Design Creationists are promoting Michael Denton's new book Evolution: Still a Theory in Crisis. The new buzzword is "structuralism" and it's guaranteed to impress the creationist crowd because nobody understands what it means but it sounds very "sciency" and philosophical. Also, it's an attack on "Darwinism" and anything that refutes evolution has to be good.

You can watch Michael Denton explain structuralism ... it only takes a few minutes of your time.

As Denton says, the basic idea is that the form (structure) of modern organisms is a property of the laws of physics and chemistry and not something that evolution discovered. He would argue that if you replay the tape of life you will always get species that look pretty much like the species we see today because the basic forms (Baupläne) are the inevitable consequences of the underlying physics.

To someone like Denton, this is confirmation of his view that God created the universe and endowed it with all the properties (laws of physics and chemistry) that would inevitably produce humans.

This is an old idea. Denton talks about Richard Owen (1804-1892) but there are many other famous structuralists from the 19th and 20th centuries. It's not a kooky creationist invention.

We should be learn about the modern versions of structuralism since the arguments take into account our current understanding of biology and evolution. One of the modern proponents was Brian Goodwin (1931 - 2009) a biologist at the Open University (UK) who was associated with the Santa Fe Institute in New Mexico (USA).

Goodwin wrote How the Leopard Changed Its Spots where he outlined his structuralist views. I read this book when it first came out in 1994 and I couldn't figure out what he was talking about back then. I have a better understanding now.

Here's an explanation from Goodwin himself published on John Brockman's Edge website: Biology is just a dance.
To introduce the problem: I like to compare morphogenesis with hydrodynamics. Suppose you have a fluid, and you want to understand why it takes certain shapes and forms: wind passes over it and it goes into waves, or you get whirlpools at the bottom of waterfalls. Why do liquids take these forms? What you need is a physical theory of fluids, which are a state of organization of matter. It's the same type of problem with organisms. Organisms are states of organization of matter. There are certain principles of spatial order in organisms, in cells, in the way cells interact with one another, and these can be written down as rules or equations, and then you can solve the equations on a computer and find out what shapes emerge, exactly the same way you can with liquids.

The hypothesis here is that life is a particular state of organization, a physical and chemical system. The problem is to find out what the rules are that apply to this state of organization of living systems.

So I see myself more as a physicist than an engineer, involved in a new synthesis of physics and biology. It's been attempted before, most notably by the Scottish zoologist D'Arcy Thompson, in his book On Growth and Form, in 1917 — an amazing achievement. He single-handedly defined the problem of biological form in mathematical terms. It's changed now, because we have new mathematical tools and a lot of new knowledge about organisms.
I want to emphasize that most structuralists do not connect their ideas to gods or creationism. Structuralism is anti-evolution but only in the sense that it requires something ("form") that has to be added to evolutionary theory ("functionalism") in order to explain the history of life.

The Edge published comments on the Goodwin article. People like Richard Dawkins, Stephen Jay Gould, Lynn Margulis, Dan Dennett, Steve Jones, and Murray Gell-mann all express their skepticism.

I particularly like the comment from Steve Jones ...
I've read some of Brian Goodwin's stuff, and I find it extremely hard to follow. That could be because I'm stupid. But I did embryology, I did development, I read molecular biology; that isn't so difficult to follow. Goodwin makes it hard. It's not an approach I like. I think he's a mystic. Anybody who goes to Santa Fe — there's something in the air there that's catching. Complexity is catching, that's the trouble.
I've been to Santa Fe and I spent a month at nearby Los Alamos. It's not the air that turns you into a fan of complexity and mysticism, it's the people at the Santa Fe Institute.

Denton has a paper on structuralism published in Bio-complexity a few years ago (Denton, 2013). The abstract does a good job of explaining the structuralist position.
Here I first review the structuralist or typological world view of pre-1859 biology, and the concept that the basic forms of the natural world--the Types--are immanent in nature, and determined by a set of special natural biological laws, the so called ‘laws of form’. I show that this conception was not based, as Darwinists often claim, on a priori philosophical belief in Platonic concepts, but rather upon the empirical finding that a vast amount of biological complexity, including the deep homologies which define the taxa of the natural system, appears to be of an abstract, non-adaptive nature that is sometimes of a strikingly numerical and geometric character. In addition, these Types exhibit an extraordinary robustness and stability, having in many instances remained invariant in diverse lineages for hundreds of millions of years. Second, I show that neither Darwinism nor any subsequent functionalist theory has ever provided a convincing adaptive or functionalist explanation for the Types or deep homologies. Third, I discuss how recent advances have provided new support for the structuralist notion that the basic forms of life are immanent in nature. These include the discovery of the cosmic fine-tuning of the laws of nature for life as it exists on earth, and advances in areas of molecular and cellular biology, where it is apparent that a considerable amount of biological complexity is clearly determined by the self-organizing properties of particular categories of matter, rather than being specified in detail in a genetic blueprint as functionalism demands.
One of the problems with structuralist explanations is that they are very animal-centric. They deal almost exclusively with examples from that tiny little circle (right) on the tree of life.

It's easy to think of deterministic forms when your view is limited to vertebrates and other animals but it's much harder to defend structuralism when you're comparing bumblebees and mushrooms.

There are plenty of good critiques of structuralism in the scientific literature. One of the best is by Paul Griffiths, an Australian philosopher of science and colleague of John Wilkins (Griffiths, 1994). Structuralism is a challenge to classification and definitions of species. I'm hoping that John will help us understand structuralism.

Here's what Paul Griffiths says,
The generic forms divide the overall space of biological possibility into discrete regions available to each particular type of organism. The process structuralists view this structuring of the space of biological possibility as part of the fundamental physical structure of nature. But the phenomena of phylogenetic inertia and developmental constraint do not support this interpretation. These phenomena show that the evolutionary pathways available to an organism are a function of the developmental structure of the organism. However, nothing in the phenomena suggests the sort of manageable periodic table of organismic forms that would be necessary to make the structural explanations of form envisaged by Goodwin (1994) genuinely explanatory. The generic forms that exist in nature may be a tiny subset of the possible generic forms that could have been created by the historical design of alternative developmental systems. In that case, an explanation of the organism's form in terms of which developmental system it possesses would in no way displace the Darwinian explanation of form in terms of descent with modification. The developmental system could have been any one of a number of ways depending on the particulars of evolutionary history.
There's nothing in science that supports the views of the structuralists. We have perfectly good explanations for why bumblebees are different than mushrooms and why all vertebrates have vertebrae and not exoskeletons. There's no evidence to support the idea that if you replay the tape of life it will come out looking anything like what we see today.

You can be confident that when you visit another planet you will not find vertebrates.1 Denton is wrong.

1. Or God.

Denton, M.J. (2013) The Types: A Persistent Structuralist Challenge to Darwinian Pan_Selectionism. Bio-Complexity 3: 1- 17 [doi: 10.5048/BIO-C.2013.3]

Griffiths, P.E. (1996) Darwinism, process structuralism, and natural kinds. Philosophy of Science, S1-S9. [PDF]


Dave Carlson said...

It's interesting that SJ Gould was critical of Goodwin's ideas. Wasn't Gould at least somewhat sympathetic the structuralist outlook on the history of life? It's been about a decade since I read The Structure of Evolutionary theory, but that's the impression I got.

Jonathan Badger said...

There's nothing in science that supports the views of the structuralists

Are you discounting even D'Arcy Thompson's "On Growth and Form"? There was quite a lot of good stuff in there, and even though he and J.B.S Haldane butted heads at times, it is obvious that Haldane couldn't have written his "On Being the Right Size" without it.

Petrushka said...

I'm probably just stupid, but I see no way of mapping changes to the genome to structural changes in the organism. Can anyone, following first principles and not copying existing sequences, design a change to an arm or leg, or any other structure?

I don't seen any evidence of a grammar or syntax to regulatory sequences. I see no way of drawing blueprints in DNA.

Joe Felsenstein said...

One of the big arguments about morphological evolution is between morphological systematists and quantitative geneticists. The quantitative geneticists tend to argue that one can select form in any direction and have response to the selection. The morphological systematists tend to view the possible changes of form as confined to rather narrow channels. Gould argued the latter view, which is where the "sructuralists" are too.

I think that the interesting issue is not whether there are such constraints -- surely there must be at least some -- but whether at present we can find out anything very interesting or general about them.

Alex SL said...

This was very enlightening. Some systematists who still want to formally recognise paraphyletic taxa have recently started to claim that the current mainstream view of accepting only monophyletic taxa is "structuralist" (e.g., and I did not even understand what that means. As far as I can tell, their definition must be different from the one presented in the post though, because under the latter the claim doesn't seem to make sense...

Aceofspades said...

Lookup HOX genes

Alex SL said...

Also, it seems to me as if there is a really simple explanation for the observations justifying structuralism as defined here: developmental constraints.

As for whether we will find vertebrates on other planets I guess that depends on how to define vertebrate. If it is "heterotrophic mobile organism with a jointed internal skeleton" I don't really see why that shouldn't pop up on many other planets given enough time, even if the details may differ (maybe some of them have their version of a spine on the belly, or ten legs, etc.). There is a limited set of things that work, after all.

Petrushka said...

First, that's copying an existing sequence. Second, there's no design toolkit possible to modify structures to order. There may be a few mutations or alleles known, but that's copying.

If I'm not mistaken, Denton and IDists are implying that some entity makes things like HOX genes from first principles.

Petrushka said...

Or, perhaps HOX genes are inevitable, destined. Holes in phase space into which sequences will fall by gravity.

Larry Moran said...

I'm probably just stupid, but I see no way of mapping changes to the genome to structural changes in the organism.

Really? You think we don't known which genetic changes are responsible for blue eyes and brown eyes in humans?

Every one of Mendel's seven pea traits involved structural changes and we know exactly what genetic changes are responsible for four of them. For example, the mutation affecting the height of the plants (tallness or stem length) is in a gene required for the synthesis of gibberellin GA1—a growth hormone. The "tall" variety carries an allele that makes 10X more gibberellin than the short variety.

Petrushka said...

But can you make a new allele and know from first principles what it will do? I have no doubt we have a growing library of alleles, but how big is our library compared to the space of possible libraries? Can we design a useful medical molecule without trial and error?

Jmac said...

Is Denton actually saying that the laws of nature were fine-tuned in such a way to produce life first (whatever that means today)and then the simple life forms lead to human because of the laws of nature?

Is Denton actually saying that the same fine-tuned laws anywhere in the universe would produce the same results?

Can someone either confirm or deny any of my conclusions whether right or wrong? Please!

John Harshman said...

Does Denton identify any of these "types"? Are there types within types or are all the types disjunct? Presumably, since Denton accepts universal common descent, these types have evolved at some point in response to the push of whatever physical laws he supposes. Now why should different species, starting from the same point, evolve different types? Are physical laws different in different places? I confess a near-total lack of understanding.

Petrushka said...

"There is a limited set of things that work, after all."

Is there? I would say that physics constrains things that work, but doesn't define what is possible.

Robert Byers said...

Thats interesting that this has come up inn opposition to evolution.
As a YEC I noted , in the fossil record, there were limitations and constraints of options for morphological results for creatures that did change.
I remember , how a running rhino of n America (fossil), had very similiar legs to a horse. So physics is and must constraining options for body plan changes. in fact evolutionism should embrace that. gravity constrains options for evolution surely. Its messy about mutations being also constrained but there it is.

i guess Denton is seeing the Constraints as nullifying evolutions claims to control everything by mutations. therefore he might be suggesting its not just a correction of evolution but a overthrow.
Indeed however body plans do end up with the like results in nature.
there is not the differences that there should be where options can be anything.
Denton is probably at least making a inportant minor correction. I'm not sure its a great knockout however.

Dave Carlson said...

I see. Thanks for the information!

SRM said...

It seems two different questions: what was possible and what was inevitable on this planet? Seems to me there is an extraordinary amount of phenotypic diversity on this planet, but whether it is a fair sampling of most possibilities I do not know. Since life emerges from physical and chemical principles, no doubt there are constraints of some sort.

On the other hand, if an asteroid hit this earth 10 million years ago instead of 65 million years ago, I for one would not be sitting here pondering why the notion of structuralism would be considered a challenge to evolution and its study at all, nor why it would be evidence for god in the slightest.

Schenck said...

"how a running rhino of n America (fossil), had very similiar legs to a horse. So physics is and must constraining options for body plan changes"

Yes, that's called "the environment", it's that thing that an organism adapts to.

Also rhinos and horses descend from the same ancestor, so homologous structures, doing the same task, in the same environment, probably should look reminiscent of each other.

judmarc said...

Joe F, I remember a discussion of Gould's interest in this question at the end of Wonderful Life.

In my naive view the fact that key developmental genes are at the center of so much that is morphological might make it difficult to monkey too much with morphology except in some fairly specific ways (duplication of developmental genes followed by co-optation, for example).

gnomon said...

Joe F.

Birds genomes are smaller than other vertebrates, such as reptiles, mammals, amphibians, bony fish etc. There is also size reduction in non-coding DNA. I learned this from a talk at the PAG 2016 meeting by Erich Jarvis, the main leader of the birds genome project. It surely looks to me like constraints by gravity laws.

gnomon said...

In this newly published paper of mine on the old riddle of genetic diversity, also a preprint of it here, I mentioned a number of constraints on DNA variation besides complex physiology and function as in the following, including protein mass.
“Functional constraint on sequence divergence has long been appreciated, including protein structure and function, genome-wide constraints such as fold pressure (nucleic acid stem-loop extrusion pressure) and GC-pressure (the pressure for a certain base composition), and purine loading pressure (purine rich mRNA synonymous strands), mtDNA and nuclear genome compatibility, Donnan equilibrium, aggregation pressure (crowded cytosol), and protein mass/volume/heat [18,77,82-85].”

Chris B said...

If life is that deterministic, haven't Lenski et al.'s experiments pretty much refuted this hypothesis? They had about a dozen parallel lines of evolution, none of which turned out the same (although there were some similarities). Given the uniformity of environmental conditions these lines experienced, the god-given laws of structuralism should have made short work of them and produced identical, or nearly so, paths of evolution.

If that is not enough, this hypothesis could be formally tested. There is an archive of these independent replicates at various generation times that could be 'restarted'. Michael Denton must have predictions of what would be observed based on divine structuralism, which could be compared to predictions based on theoretical population genetics.

One cannot fall back on the lazy argument that 'they're all still bacteria'. If structuralism is valid, replicates in virtually identical environments should follow the same deterministic pathway, mutations and all.

How soon before the caveats and goalpost-moving begins? That may be better explained by structuralism than is evolution.

Alex SL said...

I don't understand your reply. Yes, it is precisely the physical environment (including biotic interactions) that constrains; and constraints constraint what is possible.

Steve said...

There is nothing in darwinism that says mammals should be symmetrical, or that they have four limbs.

Or that flying things can only be so heavy.

Or that swimming things will sink if they stop swimming.

Structuralism say life has taken to heart the limits of physics and chemistry and has designed stuff accordingly.

Faizal Ali said...

Yah. 'Cuz if living things could defy the laws of physics, the laws of physics would still exist and this totally would not be cited by religionists as evidence of God.

Good thinking, Steve!

Unknown said...

I think this recent paper about some of the experimental evidence might be of interest-

"In other words, although multiple evolutionary trajectories are often accessible, evolution is strongly constrained and the part of the fitness landscape available for exploration is highly variable but typically small. Thus, if we actually could replay the tape of evolution, the outcome could have been considerably more similar to the existing diversity of life forms than Gould expected."

Petrushka said...

There are constraints on language, assuming alphabets and deep grammar, but how many different languages are possible?
I have no idea whether vertebrateness derives from some physical constraints. If so, any re-running of evolution might produce them.
But they might be an accident of history, a unique event in the universe.

Larry Moran said...

@Jack Jackson

Thanks for the reference. I had not seen that paper. I will blog about it.

Dazz said...

Great stuff!, I've been asking about it for a while (replaying the tape of life, fascinating topic)

Anonymous said...

While Owen (and Denton) are certainly wrong about vertebrates being on other planets I think it likely that we'd find simpler body plans in common with other worlds. The 'worm' body plan is pretty basic and found in several animal phyla including chordates (snakes) so even if alien creatures had an utterly different biochemistry there still might be some that looked superficially like worms.
Where the structuralists go off the rails is where they claim that these patterns are due to universal mystical forms. It sort of reminds me of Rupert Sheldrakes 'morphogenic fields' The forms of living things are explainable as evolution constrained by chemical and physical laws.

S Johnson said...

Is this controversy really so different from previous discussions about the prevalence of convergent evolution? We know that is a common phenomenon. Witness for a single example the similarities between placental and marsupial wolves. But we can also see kangaroos as the heaviest grazing animals, rather than something resembling a cow.

Convergence between the shapes of fish and cetaceans seems to be more biomechanical while the structuralists seem to be more impressed by biochemistry. But even if you think that the properties of carbon means alien life must use them, the properties of oxygen don't drive all life forms to eschew sulfur. It's not clear the structuralists are looking at all the evidence (as the little circle in the diagram suggests so strongly.)

But I thought Gould strongly emphasized historical constraints, rather than biomechanical or biochemical ones?

Argon said...

You're right, there is something 'Sheldrakian' about it although Rupert's forms were mutable. At least his 'morphic fields' concept was more concrete. Wrong, but at least concrete.

I had the impression with Denton that not only were large-scale morphologies expected to be typologically directed by nature but that the same 'typologies' would also manifest at the molecular level. So, not only would Denton claim that 'nature' was directed to produce humans with five-fingered hands but that the genetics, proteins and modes of expression of that trait would necessarily be the same as we find today. He may propose for example that the sequence of a gene for a skin protein in a human was specifically pre-planned to arise with just that sequence and to be distinctly 'better' in humans than perhaps the corresponding gene in a chimp. At least that was my impression from his earlier 'Nature's Destiny'. Perhaps he's moved from that since then.

judmarc said...

I will blog about it.

It's a very interesting paper, so I will look forward to that.

christine janis said...

"I remember , how a running rhino of n America (fossil), had very similiar legs to a horse."

Not surprising that the limbs of the three-toed rhinos of the time looked rather like three-toed horses at the time. They're both perissodactyls and closely related. But no rhino ever became monodactyl like a modern horse.

Selection for more efficient locomotion is going to lead to longer legs. But note that horse and camel, or camel and antelope, all have long legs but the legs look very different in each case. Why do you imagine that is the case?

"Witness for a single example the similarities between placental and marsupial wolves."

Witness the scientific papers that discuss the differences. Much weaker jaws in the thylacine and much less specialized legs.

"But we can also see kangaroos as the heaviest grazing animals, rather than something resembling a cow."

You're forgetting about the recently extinct diprotodonts.

Unknown said...

@Joe: I think the general rule for constraints is that they are all "frozen accidents". If the key question to be addressed is what maintains particular non-adaptive traits throughout phylogenies (and most good apomorphies certainly fit that bill) then it is simple to note that traits evolve in the context of organisms that already have some of these arbitrary traits (for instance a crustacean might have some number of thoracal segments). And the new traits may depend on the preexisting arbitrary traits, so that changes in the arbitrary traits affect function. Hence you have some diversity in the number of thoracal segments in crustaceans, but in hexapods there are always 3.

Faizal Ali said...

It seems to me that "structuralism" could just as justifiably be invoked to explain the remarkable fact that mountains are almost invariably designed so that they are narrow at the top and wide at the bottom. If it was the other way around, they would fall over. Surely that could not be a coincidence, could it?

Schenck said...

(lipid based) Cells are round because of physical constraints. You can probably except roundish cells in many places where you find lipid-based biological cells.
Pretty much everything else is a accident.

Aceofspades said...

If it's any consolation, Michael Behe makes the same claim. Her believes that all the intelligent designing was done up front by setting precise values for the fundamental constants.

A major problem is that I don't believe quantum mechanics allows for an outcome that is precisely determined based on initial conditions.

Alex SL said...

Ah, so the gimmick is just equivocation on the term possible. Yes, there are two levels of constraints here:

Some organisms are "possible", but only for about a few minutes because then they die. Sure we can imagine a tree without organs to take up water, but it won't be viable. Sure we can imagine a land-living 50 m diameter slug, but it will perish rather horribly very quickly. I would say these organisms are not possible.

Some organisms we can imagine are "possible" but so impractical that there should be a massive selective pressure to find a better solution, and then they get wiped out by whoever has found that better solution. Imagine a human with the head pointing backwards, so that they can't see where they are walking. Absurd? Yes, but that's my point. We wouldn't expect to find a planet with an organism like that.

As for an internal skeleton being a unique event in the universe, when I look at the very first worm that had a stabilising structure along the inside of its back, I conclude two things: first, this is not really hard to get to from being a worm that doesn't have it. Second, the next step isn't hard to get to either... and so on.

I guess the different perspectives here are just whether it is advantageous enough to have a trait that it will spread, or whether it is totally random what survives and what doesn't. The many cases of convergence seem to tell me that the latter isn't the case.

aljones909 said...

"I remember , how a running rhino of n America (fossil), had very similiar legs to a horse.". The mammalian skeleton is remarkably similar across all species. The last thing you'd expect from a good designer. Compare a 10g flying mammal and a 70kg bipedal mammal (a weight difference 0f 7000x. There's a remarkable correspondence of bones from one skeleton to the other. All the favorites are there: clavicle, scapula, humerus, radius, ulna, femur, fibula, tibia. Then there's the thumb and four insanely elongated fingers hidden away in the bat's wing.
There's a photo her (if the link works):

aljones909 said...

"There is nothing in darwinism that says mammals should be symmetrical, or that they have four limbs." Not sure why you're restricting it to mammals but there are asymmetrical features in whales. Other non mammal species have much greater asymmetries. A symmetrical body shape will (in general) be more efficient. e.g. both eyes the same for 3D vision, similarly for ears, running very difficult if you're not symmetrical.

"Or that flying things can only be so heavy."
This is just physics. The heaviest flying bird today is 18kg. It's a struggle to get that weight off the ground. Watch a swan take off then watch how easy it is for a sparrow.

"Or that swimming things will sink if they stop swimming."

No they don't. Bony fish alter their buoyancy using swim bladders (interestingly, it's the swim bladder, not the gills, that evolved into lungs).

John Harshman said...

interestingly, it's the swim bladder, not the gills, that evolved into lungs

Other way around: lungs evolved into swim bladders.

Robert Byers said...

Robert Schenck
It says little about mechanism to say envirorment. Creationists say that too!
It wouldn't matter claim common descent fopr horses/rhinos about leg types. the leg type of the rhino i mentioned was not like other rhinos but like horses.
Anyways i'm going from memory about the running rhino.

Robert Byers said...

S johnson. Maybe thats point for Denton. In convergent evolution claims it is the great structure that is the dominant point. In the water there are constraints on options and so everybody must admit these constraints are in operation. In fact what else could marine creatures do but the certain types of body planns they have for a unique water world.

I don't agree marsupials are convergent but instead they are the same creatures ass placentals with adaption upon migration to certain areas. Just like in the water.

Jmac said...

John Harshman,

Oh yeah? In what steps exactly did it happen? Are you an expert in respirology all of the sudden?

Robert Byers said...

C.M. Janis
I'm going from memory. I understood there was a diversity of rhinos on the N american plains and one of them was a running rhino with, I think, legs like a horse or close. it would not be because of claimed like descent. It would be either rhinos had these horse legs and then lost them or they gained them upon need. Yet not about common descent.
The rhino leg looking like a horse leg would be from need for speed, endurance, and carrying a big bulk. in other words the horse leg is not a horse leg but just the physicsbiology constraint on options.
The horse just has a horse leg too in other words.
I suspect that because the horse did not originally have its body type when on the ark. It adapted to herd running.

A note about marsupial wolves. its not that they have weaker jawss but other wolves have stronger ones. jaw size is trivial and not worthy to ignore the vast number of anatomical points of likeness between the two.
likewise the legs. its within any spectrum of any other creatures these tiny differences .
Yes to mechanisms to change body plans a little, as in people, but not the impossible. Yes marsupials are just placentals cousins once or twice removed.

Robert Byers said...

Its not THAT similiar. be definition, of evolution too, its adapted to its needs. The running rhino leg must life a heavy creature and sppeed it along. It did not originally have that leg type. The horse also had this experience and both came to the same conclusion based on physicsbiology constraints of options.
Evolutionism has denied the importance of physics in controlling results. Its all about random mutations can do anything. Denton smells this out.

John Harshman said...

Is there even such a thing as "respirology"?

It's a simple inference from phylogeny. See for example figure 2 in this book.

Faizal Ali said...

Respirology is a medical specialty, but its practitioners will generally know very little about the evolution of the respiratory system. More than Eric does, though, of course.

judmarc said...

Oh yeah?

Eric, your keen mastery of the parlance of scientific discussion never ceases to amaze.

aljones909 said...

I didn't know that primitive fish had "lungs". My limited reading suggests these organs had a dual function: gaseous exchange and buoyancy control.

rich lawler said...

I also recall that Gould was sympathetic to structuralism (at least as it pertained to development). He (and Lewontin) first brought it up in the Spandrels paper, citing the work of Adolph Seilacher.

rich lawler said...

Gould's early work on allometry (1966) also also draws heavily from Thompson's book as well.

Robert Byers said...

Another point would be if the moon had perfect air for life and life evolved there.
The moon having less gravity would mean elephants would not be as big because they would not need so much bulk in legs etc to hold up their bodies. Or deer would not be like ours because they would not need so much power to leap up and can't have the same power because of delay in landing and so a predator catch them. So its impossible deers would look like ours if evolution is going on.
So structure must be beyond random mutation based on physics.
Structure is independent of genes doing anything. There are constraints/boundaries to options.
I don't think evolutionism has understood the physics issues with the options from gene mutations and so Denton is making a new important point.

Schenck said...

This is ridiculous. The environment is the very thing organisms are adapting to; that hasn't been missed.

Faizal Ali said...

Robert Schenck, meet Robert Byers. If you have not made acquaintance with Mr. Byers and his peculiar version of what he considers to be "thinking", you're in for a treat. This was extremely rational and coherent by his usual standards.

Robert Byers said...

R Schenck
Denton is , at least one point, saying that the structures in nature are beyond the probability IF random mutations were the driving engine.
There is more common structures then random mutations would allow.
So I add physics has been missed by evolutionary thought a little bit.
I never see them talk about gravity's control on results in the case I made.
Yes they are aware of this concept but a carelessness. at least, kicks around.
The environment is not other concepts like physics.

Its also unlikely that physic controls would lead to like results from a control on mutations being selected on.
So structure hints at evolutions unlikelyness.
its close reasoning but makes sense.

un said...

The best discussion of structuralism by Gould took place in his essay Eight Little Piggies. In that essay he seemed to embrace the view summarized by Simon in the comment above.

Aaron Baldwin said...

Simon - a great example of this constraint can be seen in Malacostracan crustaceans (amphipods, isopods, crabs, lobsters, shrimp, etc.)Despite the huge diversity of form all of the Eumalacostraca have 19 somites (loosely segments), 5 on the head, 8 on the thorax, and 6 abdominal. In addition they all have gonopores on the 6th (female) or 8th (male) thoracic appendages. It's a beautiful pattern, but no indication of any divine predestination.

PZ Myers said...

Now why would anyone trust Denton or a creationist to understand what structuralism is all about? What he says is nonsense. That isn't the structuralism I was weaned on at all.

I also don't get how you can say it's animal-centric. The best examples of structuralism are from single-celled organisms -- there were dead-end efforts to apply structuralist principles to animal development and patterning, especially in the 1950s, and they didn't work…or, that is, in theory they might have worked, but molecular genetics worked much better.

Structuralism is more or less dead right now, except as an explanation we take completely for granted in many fundamental biological phenomena, but there are still some really good, unanswered questions it brings up.

Unknown said...

I don’t care about Denton and IDiots (I’m French). But I think that the main message of “structuralists” is quite simple. Genes do not make forms, but act as parametric modulators of self-organizing systems. Physical processes are at work in development and if one wants to understand the evolution of form, one must understand the physical processes involved in morphogenesis. Myers talks about “The structuralist heresy”. But if you want a real heresy, just read this paper:
“In 1980, when biophysicist George Oster submitted a paper to an esteemed developmental biology journal, he received a reply with a rather Orwellian overtone. “The authors are attempting to apply Newton’s laws to embryos, ” the anonymous reviewer wrote, “but as all biologists know, biological systems don’t obey the laws of physics.” Luckily, an astute editor found that remark suspicious, and the paper, coauthored with Garry Odell and Beth Burnside, was published (1). “But that was the atmosphere of the day, ” Oster says.”
A real heresy of the 80 : “as all biologists know, biological systems don’t obey the laws of physics”. Just remember that the biophysicist Oster is a father of Evo-Devo, with Alberch.
Many peoples that are working on the physical basis of morphogenesis are not stamped “structuralists”. Yet for those that are involved in such studies, the message of Brian Goodwin is quite simple! A real problem is that biologists do not like maths or physics, and most of them think that “genes control” and “genes specify” and so on…? What this mean exactly? Just few recent ref (not from IDiots or from “heretic” journals!) to show that “structuralism” (if you like stamps) is not dead at all (and is not dead with d’Arcy Thompson!!!!):
Eiraku M., Takata N., Ishibashi H., Kawada M., Sakakura E., Okuda S., Sekiguchi K, Adachi T. & Sasai Y. (2011). Self-organizing optic-cup morphogenesis in three-dimensional culture. Nature, 472(7341), 51-56.
Savin, T., Kurpios, N. A., Shyer, A. E., Florescu, P., Liang, H., Mahadevan, L., & Tabin, C. J. (2011). On the growth and form of the gut. Nature, 476(7358), 57-62.
Takigawa‐Imamura H., Morita R., Iwaki T., Tsuji T. & Yoshikawa K. (2015). Tooth germ invagination from cell‐cell interaction: Working hypothesis on mechanical Instability. Journal of Theoretical Biology, 382, 284‐291.
Tallinen T., Chung J. Y., Rousseau F., Girard N., Lefèvre J. & Mahadevan L. (2016). On the growth and form of cortical convolutions. Nature Physics.
Varner V. D., Gleghorn J. P., Miller E., Radisky D. C. & Nelson, C. M. (2015). Mechanically patterning the embryonic airway epithelium. Proceedings of the National Academy of Sciences of the USA, 112, pp. 9230‐9235.

Diogenes said...

Novalice says A real problem is that biologists do not like maths or physics, and most of them think that “genes control” and “genes specify” and so on…?

Actually, *structuralists* don't like math or physics. None of them invoke the laws of physics to explain structuralism. It's mostly them making a catalog of diagrams, the biological equivalent of a diary of snowflakes, combined with a bunch of dead German philosophy about Gestalten und die mystical cosmic Einheit und Holismus und Alexander von Humboldt, und after reading a bit of their dead philosophy, I want to put on a pointy helmet and invade Poland.

They may insert some jibber jabber in there about the "golden mean", but is that really math when you're not deriving any equations?

But since "structuralists" allegedly like math and physics so much, let's call your bluff.

Please explain where Denton (or any other pseudoscientist "Gestalt" archeo-German crackpot) uses math and physics to explain why tetrapods have to have a "pentadactyl" hand, starting from just math and physics. As opposed to, say, 3 fingers or 7. Go.

Then next copy and past their mathy, physics explanation for why early amphibians have six or seven digits, and mosasaurs have varying numbers of digits like 6 to 10, and an allosaurus has 3 fingers, while a T. Rex has two. From math and physics alone. Since "structuralists" are so good at it. Go.

I like the math and physics. I do. So copy and paste the equations. I'll tell you if they're any good.


Unknown said...

Diogenes said...

If you all would like a dose of this German "Gestalten" ("typology") philosophy in the raw, here's a chapter from a book of philosophy by Hitler's old mentor, the nutty, fanatical anti-Darwinist, Houston Stewart Chamberlain.

H. S. Chamberlain on Plato, for Gestalten and against Darwin

Unknown said...

Just remebre on point. I don’t care about Denton and IDiots

Diogenes said...

I have followed your link, and strangely found no equations there. Huh.

Unknown said...


Unknown said...

Hi Diogenes
I'm talking about Brian Goodwin, not about Hitler and the Godwin's_law's_law

Diogenes said...

My mistake, I went to the wrong link *blushes.* There is indeed math on that page, although the math refers to feedback loops in genetic expression, and morphogenetic fields, not actually structuralism. The section on structuralism just tells us things like "To Goodwin, many patterns that we observe in nature are a byproduct of constraints imposed by complexity. The limited repertoire of motifs observed in the spatial organization of plants and animals (at some scales) would be, in Goodwin's opinion, a fingerprint of the role played by such constraints."

Obviously it is unwise to judge a man's work from his Wikipedia page, but this doesn't address, and seems to confirm, my general objection to structuralism, which is, that it looks like circular logic. You see that in nature there are forms X1, X2, X3 and then you hypothesize that unknown laws of physics forbid anything else besides X1, X2 and X3. Well, that's circular logic, how can it be falsified? If tomorrow we observe form X4, is the hypothesis falsified?

I fear the circular logic that says, "Everything we observe are the only things we can observe." If we observe something new and different tomorrow, doesn't that just change the set 'everything we observe', thus making the claim unfalsifiable?

The structuralism hypotheses has to be just mathematical enough to make a list of all the forms that aren't possible, by an objective means, something better than 'that's all we observed up till now.'

In practice, we've found, both among living species and among extinct fossils, many intermediate forms which would be designated as forbidden by structuralism, if you assume that the forms we observe are the only ones permitted by the laws of physics.

What do you say about all these freaky things-- the amphibians with six or seven fingers, the mosasaurs with ten digits, the half-feathers and quarter-feathers we find on theropod dinosaurs, the intermediate jaw shapes in reptile-mammal transitionals, the blowholes a quarter of the way or half the way up the skull on whale transitionals, the gradually spreading vertebrae and unfusing sacrum of the same, ...not to mention just plain freaky stuff like the crests on pterosaurs or the bony horns on brontotheres?

We have intermediate forms for many such complex structures. Shouldn't structuralism forbid us from finding that stuff? If it doesn't forbid that freaky stuff, how else can it be falsified?

Diogenes said...

Novalice: I am not accusing you of being a Nazi nor an ID proponent. However, the German philosophy is directly relevant because Denton has brought it up himself-- in the past, Denton has used the word Gestalt to describe his typology, has explicitly said that he is bringing back to life a philosophy of the late 19th century, and he has claimed that this used to be the dominant hypothesis in biology. So since Denton himself made the connection, it's fair for me to address it.

I challenge Denton's false claims that this structuralism/Gestalt philosophy used to be biology's dominant hypothesis. First, it wasn't genuinely dominant, and second, it was always pseudoscientific IMHO, no matter that big shot intellectuals pushed it in late 19th/early 20th century.

Sure it was "respectable" in the sense that big shot intellectuals in German-speaking countries pushed it in the late 19th/early 20th century, but that doesn't mean it was true science, nor that it was the dominant hypothesis.

In the 19th century, I would argue, there wasn't a dominant view of biology-- the English speakers had Darwin, the French had Lamarck, while German speakers had Humboldt, Kant, Goethe and later, Haeckel (whose Kunstformen der Natur is an artistically brilliant exploration of biological form.) The idea of Gestalt was certainly inspired by big shots, but it didn't affect English language biology much, French I don't know about, so I would not agree it was the dominant old time view.

So the question that remains is: was it ever scientific? I expect its partisans to complain when I reply, no, it was always pseudoscience. How could you falsify it?

The whole idea of Gestalt IMHO is in the same category as vitalism (another anti-reductionist idea): the vitalists would say "If chemistry can't explain this phenomenon, then the vital force caused it." But that leaves you defining your hypothesis in terms of the gaps in someone else's.

Remember Huxley's criticism of vitalism, to paraphrase, 'It's like calling an invisible force 'aquosity' and then assigning all the properties of water that we don't understand to its 'aquosity' force."

Huxley's criticism of vitalism is one which structuralists have to take seriously IMHO. How is their "structuralism" or their Gestalten different from Huxley's "aquosity"?

Diogenes said...

P.S. I am certainly not invoking some shoddy logic such as, since the Nazis absorbed ideas of holism and Gestalt into their nutty philosophy, that discredits the philosophy. The Nazis attracted every bad idea to themselves like an electrically charged sock, but the fact that Nazis absorbed it doesn't make it false.

But it is a fascinating story, though, and if anybody wants the skinny on how nutty philosophers and psychoanalysts and Nazis pushed ideas of anti-reductionism, holism and Gestalt, I can heartily recommend Anne Harrington's book Re-enchanted Science: Holism in German Culture from Wilhelm II to Hitler.

Unknown said...

Hi Diogene
Thank you for your answer. Once again, I would to say that I am not ID or creationist. I’m atheist, but with Jewish origin and have not a great sympathy for Hitler! I am working in biophysics and evolution. I did not and will never read Denton, I have many more interesting things to do. Moreover, I want to point out again that “structuralist” is only a label that tends to disappear. Why? Because during the 70’s and 80’s (and before), the few authors that tried to explain biological morphogenesis in a biophysical perspective (as Goodwin) were labeled “structuralist”, the few “bizarre authors” that applied “Newton’s laws to embryos” (see my preceding post). Now, this approach has become so commonplace that the term structuralist itself disappears because not biologist could still write that “biological systems don’t obey the laws of physics”. The structuralist label tend to disappear, but not the main message: if one wants to understand the evolution of form, one must understand the physical processes involved in morphogenesis. Now if you want some ref on the biophysical basis of morphogenesis:
Do you really think that all these authors really care about the fact of being labeled “structuralist”? No, of course. But what they describe show that the ideas of authors that were labeled “structuralist” are not heresy at all. Now, you write:
“You see that in nature there are forms X1, X2, X3 and then you hypothesize that unknown laws of physics forbid anything else besides X1, X2 and X3. Well, that's circular logic, how can it be falsified? If tomorrow we observe form X4, is the hypothesis falsified?”
Could you give me only one reference in which an author hypothesis that an “unknown laws of physics forbid” a given form? If not, you are using a straw man.

Unknown said...

A last point about your hypothetical case of a structuralist hypothesizing unknown laws of physics forbiding a given form, and also your remarks below about vitalism: Alan Turing is an often cited author among “structuralists”, although Turing has never been labeled structuralist to my knowledge. But just a quote that summarize why “structuralists” like Turing (and not vitalists):
“Unless we adopt a vitalistic and teleological conception of living organisms, or make extensive use of the plea that there are important physical laws as yet undiscovered relating to the activities of organic molecules, we must envisage a living organism as a special kind of system to which the general laws of physics and chemistry apply”.
Now a quote from Goodwin:
“Will biology join up with physics, take on its flavor, have this notion of rules, organization, regularity, order? The new movement is transforming biology from a historical science, which is what it is at the moment, the objective of Darwinism being to reconstruct the history of life on Earth. Well, that's not the style of physics. Physics is about laws, the principles of organization of matter. We're doing the same thing in biology; we're looking for the principles of organization, the dynamics of the living process. Once that's understood, you're in a position to say: "Ah! History followed such and such a course in expressing and revealing the subtle order in this particular type of organization of matter we call the living state”. (Goodwin, 1995)

Unknown said...

The first quote is from Turing

John Harshman said...

The quote from Goodwin seems to be a complete denial that contingency has any role in the history of life. Do you think he would defend that claim?

Unknown said...

Hi John

Not at all!!! Structuralist never written for example that human beings were bound to appear. The only point is that biological morphogenesis obeys some rules that make biological forms partly predictable on the basis of theoretical models of morphogenesis including physical principles. That’s all. If it were not the case (a purely contingent evolution OF FORMS as in the neodarwinian framework), theoretical biology would not exist and the papers cited in a previous post would have never been written! That’s all. You should read Goodwin (among many other authors), for example:

"The contrasting emphases between evolutionary narratives and ontogenetic principle described above, and the evidence that continues to support both, calls for a resolution that depends not upon assimilation but upon an expanded conceptual framework within which both history (contingent lineages) and structure (intrinsic order) can be accommodated. A context for such an accommodation is a dynamics of complex processes which allows the study of both emergent evolutionary order and the role of contingencies on this emergence. The dynamics is the generative mechanisms of biological form (morphogenesis) and identification of the type of morphology that can arise, combined with the contingencies of genome modification and the effects of natural selection on the stability and transformation of the dynamic forms (the life cycles of different species). Such a conceptualisation of evolution might allow the two traditions of biology, involving both form and transformation, structure and process, to be reunited”.

Goodwin, B. C. (2000). The life of form. Emergent patterns of morphological transformation. Comptes Rendus de l'Académie des Sciences-Series III-Sciences de la Vie, 323(1), 15-21.

John Harshman said...

So, in the first quote Goodwin was just being hyperbolic, or perhaps his views changed between 1995 and 2000. Now it seems to me that most of the generative laws of biological form are themselves contingent, and that the greater part of the constraints in evolution are also contingent, i.e. that where you are now constrains where you can get to much more than any general laws do.

And much of theoretical biology has been done in a strictly Neodarwinian framework.

Unknown said...

“ it seems to me that most of the generative laws of biological form are themselves contingent”

It’s an absurdity! If you take for examples two recent papers (below), both are based on elasticity theory and continuum mechanics. You should send a paper to Nature or Nature physics showing that the physical laws behind their model are contingent (or are an extension of the Neodarwinian framework!) and you will win the Ig Nobel Prize. It seems that there is sometimes more obscurentist in this blog than in a church. I think that it was my last post. Bye Bye!!

Savin, T., Kurpios, N. A., Shyer, A. E., Florescu, P., Liang, H., Mahadevan, L., & Tabin, C. J. (2011). On the growth and form of the gut. Nature, 476(7358), 57-62.
Tallinen T., Chung J. Y., Rousseau F., Girard N., Lefèvre J. & Mahadevan L. (2016). On the growth and form of cortical convolutions. Nature Physics.

John Harshman said...

Ah, flouncing. Bye, then.

But in case you're still here, I wasn't talking about simple applications of physics. If that's all you're talking about, we have no argument. But I don't think physical laws are the major constraints on evolution.

gnomon said...

Thank you for bringing up these papers. Physics laws and mathematics are all based on axioms. The physical world was built by axiomatic laws. In fact, biology should reveal new axioms rather than just merely follow known axioms discovered in physics or math, since biological world is far more complex. We have proposed two in recent papers. Bottom line, genetic variation is under constraint by the organism's construction requirements. The more complex/orderly the phenotype, the more the constraint on the genotypes or building parts.

Shi Huang

John Harshman said...

I don't believe you know what "axiom" means: "a proposition that is assumed without proof for the sake of studying the consequences that follow from it".

gnomon said...

A junk yard has no constraint on its building parts whereas a computer does. A junk yard like organism has far less constraint on its building parts or genomes than a super-computer like organism. That is just common sense. And indeed, real organisms do follow such axioms. Can Darwin followers find reasons to refute that common sense? I don't think so and at least no one has tried in a formal way since that idea was first published 8 years ago.

John Harshman said...

Still don't think you know what "axiom" means. And nobody knows what your "idea" actually is, since you have never managed to explain it comprehensibly.