Dembski pwned again: ant trails and intelligent design

February 26, 2011 • 6:28 am

If you’re into ants—and who isn’t?—you can’t do better than follow biologist Alex Wild’s excellent blog Myrmecos (the study of ants is called “myrmecology”).  It’s one of the best taxon-specific blogs around.

Alex doesn’t like to deal with creationists, but made an exception when Intelligent Design (ID) advocate William Dembski started making pronouncements on ants.  Noting that ants tend to take the shortest path between colony entrances (they also do this when travelling between a colony entrance and a food source), Dembski, writing on February 18 at the ID site Uncommon Descent, pronounced this feat inexplicable by natural selection (ergo Jesus):

Now here’s an interesting twist: Colonies of ants, when they make tracks from one colony to another minimize path-length and thereby also solve the Steiner Problem (see “Ants Build Cheapest Network“). So what does this mean in evolutionary terms? In ID terms, there’s no problem — ants were designed with various capacities, and this either happens to be one of them or is one acquired through other programmed/designed capacities. On Darwinian evolutionary grounds, however, one would have to say something like the following: ants are the result of a Darwinian evolutionary process that programmed the ants with, presumably, a genetic algorithm that enables them, when put in separate colonies, to trace out paths that resolve the Steiner Problem. In other words, evolution, by some weird self-similarity, embedded an evolutionary program into the neurophysiology of the ants that enables them to solve the Steiner problem (which, presumably, gives these ants a selective advantage).

I trust good Darwinists will take this in without skipping a beat, mumbling something like “evolution sure is amazing” or “natural selection is cleverer than us.” Dispassionate minds might wonder if something deeper is at stake here.

(“Something deeper,” of course, means Jesus.)

Well, if Dembski had bothered to learn anything about ant trails (and this takes only a few minutes of Googling), he would have realized that embedded in the ants’ tiny brains is not an evolutionary algorithm for solving the Steiner problem, but a simple rule combined with a fact of chemistry: ants follow their own pheromone trails, and those pheromones are volatile.  As Wild explains, ants start out making circuitous paths, but more pheromone evaporates from the longer ones because ants take longer to traverse them while laying down their own scent.  The result is that the shortest paths wind up marked with the most pheromone, and ants follow the strongest scents.

Wild shows a nice simulation video on his site, demonstrating how, given these simple assumptions, ants wind up taking the shortest trails.

Before we say that evolution can’t explain a behavior, it behooves us to learn as much as we can about that behavior.  Dembski didn’t learn jack.  And we shouldn’t underestimate the capacity of insect brains to store complicated information, or of evolutionists to decipher how that capacity evolved.  A good example is the waggle dance of honeybees.

Alex is actually pretty soft on Dembski, for after pwning him Wild says,

In Dembski’s defense, his error is a common one. Ant societies share enough superficial similarities to human ones that the tendency to anthropomorphize is strong. It is too easy to assume ants solve complex problems the way we humans do, with smart individuals applying brainpower to puzzle them out.

I am not as forgiving.  Dembski is not just an average joe expressing bewilderment at the “swarm intelligence” of ants.  He is supposedly conversant with evolution and biology, and is making a pronouncement against evolution in a prominent place. He should have done his homework.  Thanks to Alex for correcting him, and for demonstrating the unjustified eagerness of creationists like Dembski to say “evolution couldn’t have done that.”

Dispassionate minds, indeed.

h/t: James

96 thoughts on “Dembski pwned again: ant trails and intelligent design

  1. It really is baffling, as you point out, that they think that disproving evolution infers that a magic man in the sky did it. The best they could accomplish would be to disprove evolution (HA!) and then we’d have to start over from scratch.

    1. Disprove evolution,

      “I’d like to see a tenth rate, cheap jack entertainer like you do a trick like that!”

      From ‘Bedknobs and Broomsticks’.

  2. There’s actually a heuristic technique (rather similar to genetic algorithms) known as ant colony optimisation (see which uses exactly this technique to solve problems.

    For someone like Dembski, who has published on GAs in the past to have not heard of ant colony optimisation is… well, unsurprising, given the depth of his erudition on the subject.

      1. “Of course he’s heard of it, he just knows his audience hasn’t, mostly.”

        I think that’s right. I used to know one intelligent Christian apologist who deliberately distorted and suppressed the evidence in order to make her argument sound stronger than it actually was. She thought that she was justified. It was legitimate, she thought, to use any strategy to defeat the devil, including a ‘white lie’.

        I think her ‘reasoning’ would probably proceed shakily along these lines, “Evolution has to be wrong, however there seems to be a lot of evidence for it, some of which I can’t understand well enough to argue against. If I can’t refute the evidence, I’d better distort it so that people think I’ve refuted it. It’s fair to do that because it really is wrong, I just can’t show why it’s wrong.”

      2. So what you are saying is that Dembski knows the truth about ant trails, but has decided that this truth is a bit too dangerous to be bandied about in the public sphere, and so instead thinks it best to distort it to suit his ideological ends. That sounds vaguely familiar to me from recent discussions around here.

  3. Ants come in, ants come out, you can’t explain that.

    Bill D.

    Looks like your average ant colony is cleverer than Dembski and that doesn’t surprise me.

    1. Ha! I’m reminded of something Bernd Heinrich (“Mind of the Raven”) said with regard to his ravens:

      “…I conclude that my graduate advisors had been right; perhaps one should indeed not try to study an animal cleverer than oneself.”

      Given Mr. Dembski’s comments regarding ants, I’m left unsure whether there is an appropriate member of the animal kingdom for him to study (or comment on, as in the present case)…

    2. “And we shouldn’t underestimate the capacity of insect brains to store complicated information.”

      Corollary: We shouldn’t overestimate the tiny Dembski brain’s capacity to process uncomplicated information.

  4. Congrats to Alex for making this disingenuous commenter on the ant look like a monkey’s uncle.

    Apologies for the labored play on words. It’s just one of those days.

  5. I have never understood how people like Dembski are able to maintain any credibility with anyone. It just seems so plain upon reading or listening to his arguments that he is really not very intelligent at all. His ideas and his arguments are simplistic and really rather juvenile. And his behaviour is often like that of a petulant child.

    Ants on the other hand are absolutely fascinating. Yet another example of how simple properties can yield much more complex results. Dembski has no imagination. He is so biased that he is incapable of learning.

    1. Not only that, but he didn’t even care to try to learn.

      He leapt to a magical solution without even considering the options.

      Magic is Dumbski’s default answer to everything.

    2. Dembski and his ilk are there to fill the role of a superficially credible alternative to a theologically disturbing conclusion. Along as there is someone with some sort of credentials who is making the right noises, the evolution denier can feel that he is not just an ignorant rube.

      1. MR – I believe that you are right on. ID and the DI are scienfically and theologically bankrupt, and hopefully their fiscal bankruptcy is not too far off. Credentialed handwaving is still handwaving, but the recent Penn State report re evolution and creationism in the classroom demonstrates that it is much easier to wave than to think.

    3. These folk will grasp at any seemingly authoritative straw in order to delay the day that they finally grow up, and become honest enough to admit that they have been living an embarrassingly infantile lie for their entire wasted lives.

  6. IDiot, when he sees something that looks impressive: “Wow! That’s awesome. God is great.”

    Scientist, when he sees something that looks impressive: “Wow! That’s awesome. How does it work?”

    1. That really is it in a nutshell. The puzzling part to me is how anyone can think, or feel for that matter, that your first statement is an example of a desirable and praiseworthy mindset.

      1. We don’t have any evidence whatsoever that they “think, or feel” that way.
        All we have is that they claim to agree with this bollocks.
        For all we know, they may well be mostly lying about what they really believe, perhaps as some perverted means of gaining tribal acceptance. A shibboleth, of sorts.

  7. Even without Wild’s explanation, as an abject layperson I was able to spot two really stupid things in Dembski’s blatherings:

    1) Why a genetic algorithm? Perhaps Dembski is confusing terms here, meaning that it is an algorithm coded for by genes, as opposed to an actual genetic algorithm? But then he says the whole thing about self-similarity, which makes me thing he meant exactly what he said: A true genetic algorithm, coded for with genes. But why would he just assume that? That is so odd…

    2) The Steiner problem may be NP-complete, but that doesn’t mean there aren’t plenty of good heuristic solutions. In fact, one could probably set up an experiment just right to have the ants fall into a local minimum and wind up taking a significantly longer path than optimum…

    1. Never mind any of that. Anybody who knows enough about computer science to know what a genetic algorithm is is guaranteed to know full well what recursion is, and how powerful it is at solving exactly these sorts of problems.

      It’s not necessarily fast nor efficient. But, if you can throw enough resources at it, it will solve the problem. Since the problem at hand isn’t especially hard and since the ants have vast resources at hand, the only surprise would be if the ants didn’t solve the problem.

      When it comes right down to it, that’s all that evolution itself really is: a simple recursive function: make stuff from what you have, make stuff from what you’ve made, make stuff from what you’ve made, make stuff from what you’ve made….

      Science works that way, too.



      1. BTW: Richard Feynman spent many studied months analysing this very phenomenon.
        An amateur myrmicologist as well as a genius physicist.
        I believe that it was another genius: Freeman Dyson, who observed Feynman observing the ants.

  8. So wait. Ants solve this problem by an iterative process over time whereby comparatively successful solutions appear more frequently in later iterations? And Dembski didn’t notice?

    Its like the dude has a blind spot.

  9. Dembski is not just an average joe expressing bewilderment at the “swarm intelligence” of ants. He is supposedly conversant with evolution and biology,…

    Whatever gave you that idea? Certainly nothing he’s ever written! 😉

    1. Actually, Demski is the subject of serious scientific research.

      The consensus among theoretical physicists is that there is no such thing as an actual “nothing”, as in the apologist dictum, “something can’t come from nothing.”

      However, Demski’s mind may be evidence that there is an actual nothing. Okay, it asymptotes just short, but still……

  10. People like Dumbski can maintain their credibility, even when they are pwnded time and again, because their target audience of Bible thumpers is even dumber than they are. In the land of the blind…

    1. Yes, indeed, although it doesn’t have to be true that Dembski’s readers are dumb, which is why his offense is so grievous. Lots of people, seeing the “Dr.” part in front of his name are not going to pursue the topic much further than what “Dr.” Dembski has posted.

      The intellectual crime will be that Dembski, now corrected elegantly and efficiently by Alex Wild, will not go back and make a correction to his post. Dembski will allow his errant nonsense to stand, knowing that he is effectively lying to his readers.

      1. These religious types are fundamentally dishonest. That’s why they keep trotting out the same old examples of ‘irreducible complexity’ that have been refuted long ago. That’s why intellectual con artists like William Lane Craig keep repeating the lie that there are five ‘proofs’ for the existence of god. They are comparable to mathematicians who would indoctrinate their students with the false belief that it is possible to square the circle. Such mathematicians don’t exist, of course, because in mathematics, unlike religion, truth is considered important.

        1. But by not correcting his post, he’s LYING. I’m sorry to whine like a 3 year old. He must know that he is being dishonest. How does it not bother him? How can anyone, professionally employed in the god business, knowingly lie? Why does the god business so often have so little to do with honesty or integrity?

          1. Lying for Jesus is a most ancient tradition, one that was first formalized when Eusebius stole the principle wholesale from Plato. (Eusebius, it must be noted, was the first to “discover” the amazing account of Jesus that Josephus didn’t write, centuries after Josephus died.)

            It’s actually a very similar mindset to that of the accommodationists: the goal is not the furtherance of rational evaluation of evidence, but instead winning lots of people to your team. It’s okay to tell lies (such as “irreducible complexity”) if they bring people closer to Jesus, and it’s just as okay to tell lies (such as the harmonious compatibility of prayerful faith and investigative science) if it convinces them to “believe in” Darwinism.



            1. It’s all so “ends justify means”ish–these liars will be forgiven by god for their lies, because their lying is in a bigger cause. Why do they never ask themselves why they should worship a god who needs them to lie to get more worshippers?

          2. <blockquoteHow can anyone, professionally employed in the god business, knowingly lie?That’s a good one. Thanks.

          3. How can anyone, professionally employed in the god business, knowingly lie?

            Let me fix that for you:
            “How can anyone, professionally employed in the god scam business, knowingly lie?”

            It isn’t the sheep that are liars, it is the men of cloth that pretends the emperor isn’t naked in order to be able to fleece some more. There is plenty of anecdote, but maybe not yet statistics, that many church men are irreligious but choose to lie because they are stuck.

            Of course likely most of the high up in the hierarchy made their choice not because they were afraid of the alternative but because they liked the outcome, or worse the scam itself. Dembski is likely not one of them however, he is famous for burning his bridges. In recent perspective, Dembski is the Ceauşescu/al-Gadaffi of ID – he will go down with the sinking ship because that is all that he has left himself.

      1. An amazing video. I hadn’t heard of the phenomenon, until now. Thanks!

        I’m thinking of devious ways to effect pest control. Lay down strategically-placed pheromone trails leading to circular traps.

      2. That was my immediate thought too!

        There must be ~something deeper~ to it right. God obviously must have ~designed~ that mechanism when he was having a bad day.

        The “God is a mean kid at an anthill with a magnifying glass” quote comes to mind as well, lawl thanks Bruce Almighty.

    2. Actually, I don’t think that ID’s target audience is under-educated Bible thumpers. It’s the First Things crowd. I suspect they are terribly disappointed to have made so few friends among socially conservative intellectuals.

  11. Shouldn’t there be some sort of lifetime maximum of pwnage before his PhD is revoked?

    I mean, it’s one thing to be wrong. Never was a scientific advance that occurred without plenty of ‘wrong’ happening from point A to point B.

    But willful stupidity is quite something else.

    I would think the University of Chicago would be ashamed to have him held up as one of its products of the highest level of educational competence.

    UoC is one of the best schools in the country. What happened with Dumbski? How did the system fail so completely?

    1. Well, he was competent in learning math, if not in its research AFAIU, see critique on TalkOrigins. Math has little to do with science directly. (Well, those who study dynamics may be pretty aware of physics.) On the contrary the areas obsession with platonism hinders understanding of empiricism.

      Why Dembski got his other PhD at UIC is a mystery though. Philosophy in the wrong hands may be a deadly weapon, gouging eyes out, bashing brains in and all that.

      1. Sorry.
        One cannot do any science sans mathematics.

        (I was going to employ the pun “anything that exempted math would not count”, but refrained.
        Be thankful.)

  12. When it comes right down to it, an ant colony finding the shortest distance between two points is no more remarkable than a river “finding” the path of lowest elevation. Or, for that matter, than pedestrians wearing paths through lawns that are also — surprise! — the shortest paths between popular destinations.

    This is just another variation on the “order cannot possibly under any conceivable circumstances ever spontaneously emerge from chaos” theme. Whenever I encounter it, I like to ask the god-botherer: “Who stacks the gumballs in the gumball machine?” I’ve yet to receive a coherent response.

    All hail Jesus, The Almighty Invisible Zombie Stacker of Gumballs!



  13. This reminds me of a debate PZ Myers had with another uncommondescenter, M.D. Geofff Simmons, on christian radio ( Simmons made wildly inaccurate statements about the evolution of whales and brains. PZ pwned the *#@! out of him (I almost felt sorry for the guy…almost).

    The sad thing is, is that Simmons is the author of “What Darwin Didn’t Know: A Doctor Dissects the Theory of Evolution”, and thus is likely recognized by the ID community to be an authority on the subject.

    Bloody frustrating.

  14. Dembski is evidently preaching to the choir in his article, else how could one explain the lack of ID he thus exemplifies?

    1. On another note – I remember Alex Wild – he does amazing insect photography. I used to follow his blog regularly but lost track of him. Thanks for posting this…

  15. Dembski is truly an idiot.

    And you don’t need fancy bug professors to tell you that. Soap bubbles solve the Steiner problem.

    Dembski’s explanation: intelligently designed soap bubbles:

    So what does this mean in evolutionary terms? In ID terms, there’s no problem — ants^H^H^H^H soap bubble were designed with various capacities, and this either happens to be one of them or is one acquired through other programmed/designed capacities.

    Perhaps Dembski merely failed to learn the basic variational calculus while earning his U. Chicago mathematics PhD.

    1. And they do so in an illuminating fashion. As James Sweet mentions above, NP-complete problem makes physics have trouble “calculating” the solutions too – the bubbles may take time forming and/or include extremal paths away from optimum.

      Scott Aaronson has a paper on how NP-complete Problems informs on Physical Reality:

      “Can NP-complete problems be solved efficiently in the physical universe? I survey proposals
      including soap bubbles, protein folding, quantum computing, quantum advice, quantum adiabatic algorithms, quantum-mechanical nonlinearities, hidden variables, relativistic time dilation, analog computing, Malament-Hogarth spacetimes, quantum gravity, closed timelike curves, and “anthropic computing.” The section on soap bubbles even includes some “experimental” results.

      While I do not believe that any of the proposals will let us solve NP-complete problems
      efficiently, I argue that by studying them, we can learn something not only about computation but also about physics.”

      Among other things one can derive that time travel isn’t part of physics (or all problems inclusive NP-complete would be equally trivial).

      The take away message here is that if soap bubbles “were designed with various capacities”, they would a) be “gods” as Aaronson discuss, solving all problems trivially b) we wouldn’t know anything about physics (mainly because there would be no physics without non-trivial complexity).

      “-Oh my FSM, Dembski killed Physics! Bastard!” [/South Park]

      1. It’s worse than that. That soap bubbles solve the Euclidean Steiner problem is a basic undergraduate fact (see the title of the linked paper above). It is inexcusable that Dembski, a mathematics PhD, does not know this basic fact.

        Jerry says he is “not as forgiving,” but in fact Jerry doesn’t go as hard on Dembski as he could have done for Dembski’s failure to understand a basic undergraduate fact of his professional background. After all, this is pretty simple stuff: explain why soap bubbles intersect at 120 degrees—a mathematical question. That Dr. Dembski invokes intelligent design rather than a straightforward application of the divergence theorem is what is most damning.

        1. Oh, Dembski damns himself. My take, that is closer to the ID core, is that Dembski’s ideas perverts common physics as much as the biology he likes to piss on so much. That he is refusing to invoke math is icing on the cake – but granted, delicious icing.

          1. Surely, Dembski’s problem is a sort of infantile Platonism which supposes that if some or any mathematical patterning can be observed in the world, it must proceed from, or reflect, some higher sphere.

      2. If Dumbski did Applied Math, he would have been forced to work through the soap-bubble example before being allowed a Bachelor’s Degree, for fuck’s sake.
        There is ZERO excuse for him not knowing this.

  16. When the ants go marching three by three, Hurrah! …

    This is irrefutable evidence for the Trinity. How could those atheists be so stupid not to see it?


  17. I see where they’re at it again, in Tennessee.

    Toward this end, teachers shall be permitted to help students understand, analyze, critique, and review in an objective manner the scientific strengths and scientific weaknesses of existing scientific theories covered in the course being taught.

    That is so 2008.

  18. if Dembski had bothered to learn anything about ant trails (and this takes only a few minutes of Googling), he would have realized that embedded in the ants’ tiny brains is not an evolutionary algorithm for solving the Steiner problem, but a simple rule combined with a fact of chemistry: ants follow their own pheromone trails, and those pheromones are volatile.

    Huh! Not only did I hear this explanation so long ago I can no longer date it, I have since read research pointing out that not all ant tracks are made thusly IIRC. (Evidently ants have other senses than (pheromone) smell…)

    And for Dembski to refuse acknowledging learning algorithms on both levels, of the genome (evolution) and the phenome (pheromone trail self-organization), is both hilarious and a scary show of obtuseness.

    1. Like I said, he really isn’t too bright. I’ve known a few people who have managed to earn a PHD in one field of study or another, and yet are really rather stupid. Several of them even in their field of supposed expertise. I don’t know Dembski personally of course, but he seems to be just such a person.

      1. Eh, it’s not necessarily that hard to earn a PhD – it just requires doing a bunch of lab work and writing up the results.

        But what earning a PhD is *supposed* to involve (and this doesn’t happen with every student) is developing skills in researching a subject, identifying key issues that need to be resolved, designing experiments or proofs to resolve those issues, and following through on the research. (Also helpful: organize this work efficiently, and communicate your research results clearly and concisely when done).

        Critical thinking is essential for this process, as one can not identify the key issues or design the right experiments without it. And it’s necessary for figuring out what went wrong in your experiments (and things will go wrong). I’m not sure Dembski ever really learned this, or if he did, it didn’t stick. It’s ironic that a mathematician would be so cavalier about not “checking his work”: by not researching ant trails deeply enough to see if he was making obvious, stupid assumptions. Critical thinking fail!

        1. With Dembski, though, you really can’t tell, because he’s engaged in a fraudulent enterprise. He’ll make shit up whether or not he knows better.

          I think it’s a mistake to assume these guys are dumb. Underestimating the intelligence of con artists is never a good idea.

          1. There have certainly been examples of PhD candidates in biology departments who are known creationists, just there for the credential. Faculties have trouble deciding what to do in such cases…the coursework/research is fine…

  19. “Alex is actually pretty soft on Dembski…”

    I’m not sure about soft. Subtle may be more likely.

    It is too easy to assume ants solve complex problems the way we humans do, with smart individuals applying brainpower to puzzle them out.

    Clearly, Dembski is not a smart individual applying brainpower to puzzle out complex problems.

  20. So Dembski believes that God designed ants to solve the Steiner problem. Why would He do that? Surely the ants would be better off if they just walked in a straight line from point A to point B?

    The ant navigation algorithm seems like yet another example of a flawed-but-workable trait that we see so often in nature, rather than the perfection you might expect from an omniscient and omnipotent Designer.

  21. He is supposedly conversant with evolution and biology

    No, he isn’t. It’s not the job of ID to work out the piffling details.

    And anyway, this is an information problem and dr dr Dumbski is “The Isaac Newton”* of information science, so who are we to doubt his revealed knowledge?

    *According to dr dr Dumbski.

  22. Yes, Dembski regularly lies to his readers. Fortunately his readership appears to be in the low double digits.

    ID is dead. He’s just camping on the corpse at this point.

    1. Ah, but you forget: Christianity thrives on unabashed zombie-worship. The fact that ID was stillborn is irrelevant to the 70% of Americans who insist that one or more gods is responsible for the origins of species.



      1. Ah, but you forget: Christianity thrives on unabashed zombie-worship.

        I usually enjoy your comments, and this one really brought a smile to my face.

  23. This is hilarious, because Mr. Dembski’s ant theory was actually more complicated than what really happens. He went into way too much pathetic level of detail, but yet he claims he doesn’t like pathetic levels of detail.

  24. Would it really be the end of the world of ants took the long way around anyway? So what if they take a detour once and a while. The sky isn’t going to fall if ants don’t find the shortest path. Good grief.

  25. The most interesting thing here, I think is that yes, the ants are solving the Steiner problem – but I don’t think the Steiner problem is the one the ants should “want” to solve.

    The Steiner problem, if I understand it correctly, is the problem of minimizing the total amount of path required to allow travel between all nodes on a graph. It’s a useful solution when infrastructure to support travel is expensive, but the travel itself is relatively cheap.

    This isn’t the case for the ants, though. Since they’re just travelling over bare ground and lay down their pheremone trails wherever they go anyway, the infrastructure cost is non-existent – hence the best solution for them would be to minimize the total length of each individual trip. This is easily achieved, by making straight-line paths between every node.

    So unless I’m missing something – yes, they solve *a* problem, but they’re actually solving the *wrong* problem. But, their solution is “good enough” – which is pretty much exactly what you expect from these sort of simple algorithms.

    1. Well spotted! I was so impressed the ants could do this I didn’t think about whether it was actually the right thing for them to do.

      Thinking about it, I guess each straight point-to-point path would be reinforced only by ants travelling between those two points. The Steiner solution, however, with one path leading into each nest, is reinforced by every ant travelling to and from each nest. Thus the Steiner wins over the point-to-point, even though it’s not the ideal solution.

      Let’s file this one alongside the reversed retina as cases of evolved organisms getting trapped in an excellent but not perfect part of the solution space.

  26. I just wanted to mention the phenomenon of ant circling, seriously google it – it’s damn cool. It was pointed out further up in the thread by Oldcola but I think it deserves another mention just in case it’s missed. 😀

  27. I stumbled upon this while searching “ant trails” and read through the entire page. I noticed that while the proven science here is more than reasonable, a great deal of research was left out. To be perfectly blunt, in all your condemnation of Dembski, you forgot to do your own research of his reasonings, that is: the Bible.

    Now, I realise that you will likely shred this comment, and any reference I might make, with insults, but I request that adequate research of the book defining ID, the Bible, be made.

Leave a Reply