Oh dear. Oh dear oh dear oh dear. The Discovery Institute has turned it’s puerile, feeble attention to spiders now, and it’s as pathetic as you might imagine. It’s written by a guy, Eric Cassell, who felt it important to mention that he had an undergraduate biology degree, which he earned at least four decades ago, before committing to work in — take a wild guess — engineering, of course. Much as I respect many engineers and their work, it sure tends to breed inappropriate teleological attitudes in the brains of its practitioners.
This article, The Miracle of Spiderwebs, is typical of the type. First, talk about how freakin’ complicated spider webs are, then declare that they don’t understand how they could have evolved, which therefore means it could not have evolved. Never mind that spiders have been dependent on silk for their survival for 350 million years and there has been constant selection for more effective functionality, some old guy with negligible background in biology can’t understand it! Furthermore, it can’t possibly be because he’s an ignorant ass, he has to misrepresent the current science to make his case.
Despite great effort, humans have yet to produce anything functionally equivalent to silk. Through genetic engineering, attempts have been made to duplicate it without success. The main challenge is replicating the sophisticated and information-rich protein molecules found in the silk produced by spiders and other silk-producing arthropods such as silkworms — proteins that are nearly double the size of average human proteins. Smaller proteins do not have the strength or flexibility of spider silk. Given the advanced genetic and manufacturing technologies available today, it is remarkable that spider silk still cannot be duplicated. This illustrates just how advanced the engineering design of spider silk is.
No, he’s wrong. The silk molecules tend to be long, but not particularly “information-rich”, whatever that means. They are made up of long stretches of highly repetitive amino acid sequences, interspersed with special purpose regions. The strength comes from the repetition — they fold into crystalline bricks made of beta-pleated sheets. The length of each molecule is an obstacle to synthesizing them, of course, but there’s nothing magical or incomprehensible about the genes.
Rather, what makes them difficult to replicate is the silk production apparatus. The silk gland stores a large quantity of the fluid suspension of the silk molecules, but to make a silk fiber requires passage through a long duct that physically compresses and stretches the fibers as it extracts water and acidifies the environment, causing a controlled phase transition that makes the silk molecules align and precipitate to build a continuous fiber. That’s the hard part. It’s the gene (relatively easy) plus the specific mechanical and chemical processing that makes the silk. It is not at all remarkable that we can’t duplicate that process, but there’s nothing divine about it.
But that’s all there is to this article. Gosh, it’s amazing what biology can accomplish, says the guy who thinks it’s all about engineering.
Various spiderwebs, even among spiders of the same species, are far from identical. The most obvious reason for the differences is that each is tailored to its specific location. As the Goulds explain, “Every set of initial anchor points is different; the number of radii is contingent on opportunity; the beginning of the sticky spiral depends on where the longest several radii turn out to be. In short, each web is a custom production.” The Goulds postulate that spiders have a form of mapping ability that enables them to implement general design principles in a wide variety of circumstances. This is demonstrated, for instance, by spiders successfully making repairs to damaged webs.
And…? So…? The reason every web is a custom production is that nature is not uniform. A spider has an algorithm for navigating a variable environment: an orb weaver lays down radial lines from a central point to convenient anchor points, then walks a spiral, laying down sticky threads. The process builds a recognizable, functional sheet that varies depending on structures in the neighborhood. What is the miracle here?
Providing credible evolutionary explanations for the origin of silk and web design has proven problematic. Several theories have been proposed for the origin of both, but none have been generally accepted. Biologist and spider specialist William Shear concedes that “a functional explanation for the origins of silk and the spinning habit may be impossible to achieve.” One complicating factor is that the webs of some spiders that are more distantly related are nearly identical. Shear writes, “It appears probable that several web types are the product of convergent evolution — that is, that the same web has evolved in unrelated species that have adapted to similar environmental circumstances.” But as I will argue in Chapter 6, that is an unconvincing explanation for the origin of complex programmed behaviors.
For someone cited by a creationist, WA Shear sure has a lot of publications on the paleontology and evolution of spiders and other invertebrates. It looks like his work might have been mined to extract a misleading quote, and yep, he sure has.
That brief quote is taken from an article titled, “Untangling the evolution of the web,” and surprise, surprise, surprise, the creationist has left out the bulk of the context. I know, a creationist lies about the scientific literature? Say it ain’t so! But here’s the full and accurate quote from the paper. It starts, “A functional explanation for the origins of silk and the spinning habit may be impossible to achieve”, but then there’s a goddamn COMMA.
A functional explanation for the origins of silk and the spinning habit may be impossible to achieve, but the evolution of silk-spinning organs has been studied, and debated, extensively. Revealing evidence has come from the histology of silk glands — the details of their cellular construction — and from the embryological development of the spinnerets themselves. Histological evidence allows us to draw connections, or homologies, between silk glands in different spider groups, and embryology shows clearly that the spinnerets are paired abdominal appendages, with the silk issuing from modified setae, or hairs. So much information is available on the anatomy of the spinning apparatus, in fact, that the traditional view of web evolution rests heavily on a classification derived from the form and position of spinnerets.
This is one of those things about the creationist literature that makes me thoroughly furious: they unethically misrepresent and actively distort the work published in the credible scientific literature to pretend they are scholars. They aren’t. They’re liars. And they spin their lies out into collections of self-reinforcing nonsense. I note that the creationist here seems to think that convergent evolution is somehow an unacceptable explanation for the origin of multiple kinds of webs, and cites his own goddamn book, authored by a creationist engineer with no qualifications at all in this field, and ignores the fact that his source, WA Shear, goes on to give a summary of the evolution of diverse webs. That’s not a book worth reading, obviously.
Also, that Shear article contains a simple, clear illustration of the pattern of evolution that this creationist twit thinks can’t exist.
Phylogenetic tree shows how some families of spiders may have arrived at similar and different web designs. The Mesothelae are generally believed to have evolved first from the common ancestor of all spiders, followed by the Mygalomorphae and the “true spiders”, the Araneomorphae. The monophyletic hypothesis of orb-web origin (which is incorporated into this diagram) holds that the orb-web was invented by an araneomorph, the common ancestor of araneoid and uloborid spiders, that had a cribellum. The cribellum was acquired by a spider that was the common ancestor of all araneomorphs, including the araneoid superfamily and the uloborids. The araneoids lost the cribellum, and some araneoid families later lost the orb. Among the uloborids and their close relatives, the dinopids, are many species that have modified the orb.
But wait! There’s much more, all conveniently left out of Cassell’s stupid mangling of the article. We do have good models for the evolution of spider webs, all based on, as Shear notes, the evidence from taxonomy and histology and molecular biology.
Hypothetical pathways of spider-web evolution form a tangled web of their own, with the question of the orb’s origin, and its role as a possible precursor to other webs, at the center. In several cases it’s not clear which web is ancestral; it is possible that some aerial sheet webs preceded the orb web, whereas others developed from the orb. Pathways that are less likely are indicated by light-orange arrows; for some of them there is no direct evidence.
Oh look. Someone had the proper respect for the evidence and was happy to talk about the strengths and weakness of the support for their model, and it wasn’t Eric Cassel.
I’ll also note that Cassel is relying on a generalist article from 1994 which incorporates little of the evidence from modern molecular phylogenies. He didn’t understand the 27 year old article, there’s little hope he could grasp a modern analysis, in which the details of evolution have been strengthened. Here’s an example from 2012, Early Events in the Evolution of Spider Silk Genes:
Spidroin gene tree is based on a ML analysis of the carboxy-terminal encoding region with gaps coded as binary characters and monophyly of some groups constrained (see Methods). Numbers next to nodes and terminals correspond to numbers in supplementary Tables S1 and S2 showing support values, alternate rootings, and continuous character data. Spidroins are colored according to the taxonomic group from which they were characterized: purple = Mesothelae, blue = Mygalomorphae, green = Araneomorphae. Gray squares indicate duplication events inferred by reconciliation. Hash marks on branch indicate arbitrary shortening of branch for figure quality purposes. Brackets indicate clades with the following abbreviations: AcSp = Aciniform, TuSp = Tubuliform, PySp = Pyriform, MaSp = Major ampullate, MiSp = Minor ampullate, Flag = Flagelliform.
A more fundamental challenge for those seeking to provide a detailed, causally credible explanation for the origin of silk and spiderweb architecture is the number of genes involved in producing silk, and the complex genomes of spiders. After decades of failed attempts to provide a causally adequate explanation, one can be forgiven for concluding that we have no compelling reason to assume that a step-by-step evolutionary pathway to such an information-rich substrate actually exists. And as we will discuss later, there are now some positive reasons to consider that such information-rich systems have for their source something other than a purely blind material process. Here, suffice it to say that the behaviors and functions associated with both silk and web spinning exhibit many characteristics of human engineering, and engineering of a very high order.
Repeat after me:
Complexity is not evidence of design.
Complexity is not evidence of design.
Complexity is not evidence of design.
Also repeat after me:
My ignorance is not evidence of design.
My ignorance is not evidence of design.
My ignorance is not evidence of design.
Biologists have provided causally adequate explanations for the origin and evolution of diverse spider webs, Mr Cassell is simply intentionally and maliciously ignoring them, and further, lying about the content of the scientific literature to make a claim that arachnologists are as ignorant as he is. He’s also logically contradictory.
If, Given the advanced genetic and manufacturing technologies available today, it is remarkable that spider silk still cannot be duplicated
, how can you then turn around say that webs exhibit many characteristics of human engineering, and engineering of a very high order
?
PZ, I don’t know how you do this, day after day. Sifting through this never-ending barrage of stupidity would kill me in despair.
Kudos for your hard work exposing these frauds who are almost certain to be the end of us all.
TL:DR If I don’t understand it, god did it. Such a toxically egocentric attitude.
As a totally unjustified but anecdotally supported observation, I differentiate scientists and engineers. Engineers, I see as seeing science as a collection of axioms that apply to the subjects the engineers work on as tools to create a predicted outcome. Engineers then see the world as something they control and manipulate and define. This tends to go with a more politically conservative world view. Scientists tend to not have the same sense that the world is theirs to control but rather science is pervasive to everything and their role is to understand how it behaves. If they are in an applied field like myself the goal is to figure out how to make things work despite what the world tends to do, with a sense of pushing the border of the not-doable a little further each time. So I know a few (3) Creationist or anti-vax engineers but all the scientists I know think this is insane because science applies to ourselves not just the areas the engineers works on
Engineers, particularly those with only undergraduate degrees, think they can understand anything with five minutes and a data sheet. They get a superficial smattering of basic science, just enough to be dangerous. Data sheets are expected to have errors, but not intentional lies. Those are in the marketing brochures. Following a trail to original sources is not something they are necessarily good at. This makes them prone to various forms of pseudoscience.
A particular friend spent a lot of time studying stock market chart patterns. He seems to believe they are real at some fundamental level. There have been many books written about deciphering the patterns, of which there are 100’s. He gets very annoyed when I point out that statistically the charts are indistinguishable from random walks. He brushes me off with “Well Joe Blow made a lot of money from the patterns in his book.” Maybe, maybe not. No way to tell. And that doesn’t prove anything anyway.
Years ago I saw a membership list for the Institute for Creation Science (ICR) –Henry Morris’ ripoff of George McCready Price. It had lots of engineers. It would take some amount of research to find out that Price, while he wrote well, misrepresented the real science. You have to have knowledge of geology, or dig deeper to expose the fraud. Obviously they hadn’t bothered.
I agree with the point about teleology and it explains a lot. If your habit is to evaluate any system based on the assumption that all of its parts “are there to do something” it will leave you with a poor way of understanding nature. Actually living things subject to selective pressure can give this impression, but other forms of self-organization such as the Giant’s Causeway or snowflakes have no interpretation in terms of function.
What I resent* (and I’m not accusing PZ) is the suggestion all engineers are such poor thinkers. It is possible to be both a scientist and an engineer and to know when to put which hat on. (It seems sort of clear that experimental scientists have to do some of their own engineering of equipment for instance.)
The problem is not engineering as such but the ill-placed confidence by some not all engineers that the technical methods they’ve learned in their field turn them into experts on other unrelated technical areas, and the lack of awareness that their engineering mindset is counterproductive to observing nature. Someone who is first a decent critical thinker and second chooses to pursue engineering has not taken the highway to guaranteed brain-damage. It is the lack of critical thinking rather than the presence of engineering discipline that results in the stereotypical creationist engineer.
*As a software engineer (which isn’t an engineer) and computer scientist (which isn’t really a scientist) I get a little touchy about all this. I’m a software developer trained academically in a field of applied mathematics by convention called “theoretical computer science.” But it’s not a natural science, rather the study of abstraction, and the methods used in software development are rarely as constrained as conventional engineering disciplines like building a bridge, working out a bulk chemical synthesis, or designing an electric circuit. (But I’m not proud. “Code monkey” is fine.)
Thanks PZ, I too am amazed that you take the time to read anything produced by creationists.
And bcw bcw: great explanation of engineers vs. scientists! Copied to my files with attribution.
I’m suddenly curious about caterpillar silk and spider silk. They’re both proteins, but not the same, and they seem like an example of convergent evolution (considering one is an insect, the other an arachnid, and there are not many other silk-producers). Do they have any homologous genes for silk? E.g., I could imagine some predecessor with a different role like mucus that might have evolved into in different silk-like proteins. Someone must have studied this, right?
AstroLad@4
I put on my amateur economist’s hat for claims like this. If you could prepackage chart patterns as an algorithm, then anyone could apply them in an attempt to get rich. Since every stock sale involves someone selling too low or buying too high, it follows that not everyone can get rich this way. Thus, as soon as a pattern is discovered and widely broadcast, it will stop working. (In short, the old punchline “If there was a $20 bill on the ground, somebody would have already picked it up.”)
Caveats: (a) Some people think they’re “smarter than the average bear” so not everyone knows their technical analysis (b) everyone can make money in the stock market assuming economic growth, but it’s a slower process (c) I am not an economist.
Where’s the data sheet for this spider? It reminds me a little of Richard Feynman’s anecdote about asking for a map of a cat.
bcw bcw @3
Good engineers know that control is often illusionary, and that the real world is more complicated and messier than the models they were taught. We use such models because can give a reasonable approximation of reality while remaining usable.
They are also aware of the limits of their knowledge.
Let me offer myself as a counterpoint; an engineer and a liberal (probably a flaming lefty by US standards).
Actually I have a very dim view of any political party that is driven by ideology or demagogues. Both tend to ignore reality if it doesn’t align with their views. Usually that doesn’t work well.
AstroLad@4
As an experienced engineer I can truthfully say that the more I learn, the more I am aware that I know very little.
The cherry-picking they all indulge in means not just ignoring the rest of a page or a paragraph but ending a sentence half way through or deleting some words within a sentence. So they must know that they are lying, or do they never read the second half of a sentence because the first helf agrees with their belief? Or are they so ignorant they don’t understand the data which contradicts their blind faith? It wouldn’t matter except this creationist nonsense show no sign of abating, and is indeed being reinforced by the growing proportion of children who are home-schooled or go to religious schools.
I’ve taught engineering (electrical) at a local college for 40 years. I have met good and bad engineers, and good and bad engineering professors, so let’s not be too hasty to make sweeping judgements about engineers as a group. One of the problems is that the bad ones are used to what I call “received knowledge”. For example, there is a formula in the book. They don’t care too much about where it came from, they just teach it or apply it. The good ones do care, and also accept the fact that those formulas are really just models of reality, and like all models, they have limits. This is sometimes humorously seen with circuit simulation software. I can’t count the number of times I’ve looked at a simulation result and thought “that’s not quite right…” yet others will simply claim that it must be right because that’s what the computer shows,
It should also be pointed out that a large percentage of engineers work in the defense industry and there tends to be a consistent undercurrent of conservative political thought in those areas. I imagine that it may influence the thinking of young and impressionable engineers.
Finally, many engineering students are told repeatedly that they are the “best and the brightest”. Constant praise like that can warp your self image. I knew a prof many years ago who taught an advance engineering class and he used to great the students on the first day by saying “Welcome to the hardest class on campus”. I never said anything to him (being an underling at the time) but I always thought “Oh yeah? And how well do you think you’d do if you were thrown into an advanced piano course?”
What’s my quickie method of determining the good from the bad upon first meeting? Simple. I ask what sort of hobbies they have. In my experience, if they have an avocation that involves some sort of artistic creativity (playing/composing music, painting, writing poetry, etc.), the chances are high that they will be a good engineer. If they tend to sneer at those sorts of endeavors, not so much.
sorry. great=greet in 3rd para
@9. I use the “No True Scotsman” definition of engineer – if you have an engineering degree but think like a scientist then you are a scientist with an engineering degree.
Just joining in on this piling-on of the assertion Barbecued FSM in a tutu, the cretinist-style leaps without apparent bound seem so obvious that postulating such as an theorem perhaps justifies being chucked over the edge of the Earth to count all the turtles on the way down.
This couplet from the epic poem Marmion by Walter Scott (1808) is apt here:
I worked on modelling the web-construction of a particular orb-weaver some 30 years ago. The project didn’t get as far as I would have liked, but I did conclude that the spiders did not need to have any sort of internal “map” of their web and the environment in which it was placed. Rather, they seemed to start out by attaching their trail of silk to more or less random points until something like a central area emerged, then be guided by what they had already produced – IOW, the web was its own developing “map”. The construction process became more and more stereotyped as construction proceeded, until, after a certain point, the spider would no longer attempt to repair the web (e.g. if some of the radii were burned away), but just proceed to attach first the non-sticky and then the sticky spiral as best they could. In retrospect, a more interesting project would have abstracted away some of the details, and looked at how starting with a very simple algorithm producing a near-random “tangle”, then refining it in various ways, might have led to the huge variety of webs different species produce. Maybe I’ll go back to this work…
imback@16,
I don’t recall who came up with this.
KG@17 I suspect that human beings also carry out many tasks without a mental picture of the result they’re after or a map of the environment in which they’re doing it. The basic algorithm of “keep tweaking until you like the outcome” is incredibly effective and it should not surprise anyone, though disingenuous creationists always want to claim it does.
First thing I thought of was Bilbo Baggins calling the large spiders , Attercop ! . It’s actually an old English word for spider . Tolkien used it for The Hobbit . You learn something new every day ! I feel sorry for creationists stuck in deliberate ignorance.
PaulBC@19,
Yes, most of them, I’d say! My D.Phil was on how people can find their way from place to place without advance planning, or consulting any sort of “mental map”, and the same applies to many other tasks. The other side of the coin is how much of our thinking is dependent on “cognitive prostheses” – parts of the external world that assist us in problem-solving. Andy Clark’s Supersizing the Mind explores this.