Saturday, January 07, 2012

Chris Hogue on Complexity and Evolution

Chris Hogue is a Canadian biochemist/bioinformatician who works on protein folding (among other things) at the National University of Singapore. He used to be a professor in my department here at the University of Toronto. I miss him, and wish he were still here.

Chris blogs at BioImplement and he has just started a new series of posts on Complexity and Evolution. His goal is to explain how human design can inform us about evolution. The idea is to refute the arguments of Intelligent Design Creationists who treat intelligent design as something mystical that's1 beyond naturalism.

Here's how Chris explains what's coming ...
The thread connecting these examples of human design is that each one is an analogy to biological evolution, from which evolution may be better understood by laypersons. Now by posting new examples like this, I realize that they may all be stolen by the “intelligent design” (ID) creationists to argue against evolution. My view on ID follows that most clearly expressed in the 2005 court judgment from the Pennsylvania Kitzmiller v. Dover case: “The overwhelming evidence at trial established that ID is a religious view, a mere re-labeling of creationism, and not a scientific theory.” Of course a few scientists have written in defense of evolution and against ID nonsense in the classroom, the most strident of whom is Richard Dawkins. I now add my voice in support, as in his final interview with Dawkins, Christopher Hitchens lamented “It’s the shame of your colleagues that they don’t form ranks and say, ‘Listen, we’re going to defend our colleagues from these appalling and obfuscating elements.’”

So into the breach, I add my voice with some new arguments, after this small bit of throat-clearing. I will try to avoid being derivative as I come armed with my own capacity for inquiry, insight, and argument. My examples will show how ID concepts force the gerrymandering of human design history, and surround it with mystical borders to make their claims. The individual steps in human design are small, slow and absolutely require the intellectual imprinting of lessons by trial and error. Students who are led to think falsely about human design, or any complexity as having mystical origins are harmed by the diminishment of their own aspirations of creativity. We all need to understand how small steps and tools lead to human creativity and any object of complexity. I will reveal these small steps and show, where I can, the failures that led to success.
I know Chris and I can assure you that his upcoming posts will be provocative and informative.


  1. There's certainly selection in human design--but it's not usually from random variation. That's an important difference!

    Most important advances are based on perception and insight. Even the selection is generally based on these.

  2. ID is but intelligently managed propaganda. A bullshitology machinery. Thus, whatever good intentions Chris might have, it will be no use. Actually, it will give the IDiots more ammunition for that rhetorical trick of "since this requires designers evolution requires God, I mean, an intelligent designer."

    No way out but by learning to dismantle the rhetorical trickery built by the IDiots in a way that makes it painfully evident to those who buy their snake-oil. At least evident to those who might be on their way of buying the snake oil but have not fallen for it yet.

  3. Thank you Professor Moran for the kind words and for the posting. 

    @lee_merrill - Patents, especially in early glass soda bottle closures, forced designers to come up with variations that explored a very diverse solution space using crude tools and materials at hand. Yes, not exactly random, but certainly as strange and surprising as the Cambrian explosion. Perception and insights are always limited by the human designer's knowledge and multiple failures always precede success.  I will cover how human design and market selection are separated in the next few essays as regards bottles and closure designs.

    @Negative_Entropy - futile to convert die hard ID followers - agreed. That is not my purpose.

    My lines of argument are aimed at people who are on the edge of losing their belief in ID as their last handhold on faith and for whom history is a language they can understand and a subject in which they can apply critical thinking. Unlike biochemistry, history is more broadly understood and certainly is misused by ID straw-men arguments. Independent of that group, I hope for readers who can benefit from tangible analogies to understand molecular evolution. I will not give these historical analogies up out of fear when, accurately reviewed, and in language laypersons can understand, they can help me to educate people. 

    It would be a great complement if an example of mine were stolen by ID proponents, as it proves my claim of intellectual laziness amongst their ranks, and offers an opportunity to debate. I have an enormous stockpile to draw upon and you will find a surplus of ammunition to use for yourself, but as Uncle Milty was claimed to have said - 'only pull out enough to win'.

    I hope you will stick with me, as the best is yet to come, and thank-you again for your comments.

    Christopher Hogue

  4. I read Hogue's first post and have an initial comment.
    When you say, "The individual steps in human design are small, slow and absolutely require the intellectual imprinting of lessons by trial and error," I think most everyone would agree that's true. But most if not all human design begins with a goal, an endpoint to be achieved. How one gets there will usually be in short incremental steps, but with a specific aim. Evolution's only goal is to survive and reproduce. That's not specific enough to build the complexity and the molecular level we see throughout the cell. After all we are routinely told that evolution is without plan or purpose.

    1. Ray,

      Are you sure that surviving and reproducing is not specific enough for complexity to arise? Are all environments identical? Do organisms activities modify environments at all? Do some organisms feed on others?

      It is so easy to start imagining what would happen as an organism grows and strives in an environment, some resource getting scares, mutations allowing a subpopulation to feed on another resource, some organisms starting to eat other organisms, and such ... it becomes quite complex very rapidly, and surviving and reproducing becomes quite complicated. The adaptations necessary for survival become more complex, just as organisms would become more complex as a response. All building upon previous and simpler forms ...

      Why, I wonder, is it so hard for creationists to imagine environments as a complex and dynamic problem for survival? Don't creationists get burn under the sun? Don't creationists feel the cold in winter? Don't they feed on variable organisms? Don't they fear being eaten by a tiger? Don't they fear being eaten by microbes? Don't they ...

    2. Negative Entropy,

      My answer to the first three questions is no, and my answer to the last 5 questions is yes. My point is that I don't entirely see the relevance. I've studied animal ecology at the university level and I understand the complexities of environments. Maybe my scientific imagination isn't as robust as yours. Even survival and reproduction is not a "goal" in the evolutionary system. It's simply what organisms do. Those that "happen" to survive and reproduce better are selected. Organisms don't really solve problems as we do in human design. Again, they just do what they do with what "equipment" they have.

      Human design is inherently different. We observe a problem and construct a solution which needs to be tested and refined. But each refinement is thought about. Evolutionary refinements just happen. Maybe they are selected and maybe not. Sometimes even good solutions may not make the grade due to stochastic events that wipe them out before they can be selected.

      It's the thought process that makes the difference. Perhaps that's just my personal incredulity speaking, but I don't think so. I am trying to "imagine" the evolutionary system within the system. But as I'm sure you recognize, imagining is not enough. I need to see it. Even in my own research decades ago, speciation just happened. Trying to understand the causes and relevant events is truly mind boggling. It could have simply been a stochastic event and no real problem was solved.

    3. Hey Ray,

      To be sure the metaphors must break at some point. I agree with you that there's differences, and that the thought process is one such difference. I agree that imagining is not enough. This is where we would have to go and study about such things as genetic algorithms, or about experimental/directed evolution. But they might not satisfy you. Let's try one example, genetic algorithms have been used to "design" chips that solve some very hard problems, and, once built, they work, but electronic engineers have a very hard time figuring out how they work. Thus, complexity was built by exerting a selection environment consisting on whether the chips performed some function. Yet, the background was random variation, and solutions "just" happened. The best performing were put again through random mutations, and improvements (and complexity) achieved/built upon, and so on. It works. Yet, the metaphor of human technologies would not capture this part of the evolutionary process. Success behind a random background. But the human technology metaphor might capture well the idea that even in human technologies we had to start very simple with complexity arising on top of that.

    4. N.E.,
      I don't really know much about genetic algorithms. It certainly is intriguing results. My suspicion is about the code generating the algorithm itself. Not that I suspect that there is some kind of built-in directions but simply that the undirected variations are being generated by something quite complex in itself. As you hint at above its still a metaphor. From what you reported it also seems they were looking to solve something in particular. Which of course evolution doesn't do.

    5. Yet again, evolution does solve something in particular: survival given changing environments.

      As for the code. It certainly has to have rules. For instance, you would not allow for say, semi-conductors that are not semi-conductors. But this is somewhat equivalent for there being natural laws. So, mutations and improvements in life forms do not happen by random arrangement of atoms because the living systems have a way of reproducing, which limits the ways in which mutations, recombinations, and reproduction can happen. All of this requires energy. And so on and so forth. There are complicated rules in nature, and those rules actually both limit and explain a lot about the evolutionary process. I am not talking just how living forms work today and have worked through eons, but about how nature works in general. Waves of heat and cold are not random, we have a day and night because there is rotation, and the rhythms have consequences that can produce some pretty complex patterns in winds, then in crystallization, then in ... maybe you get the picture. Nothing in nature occurs in a rule-less void. I just saw a documentary about a place in Sudbury, Ontario (if I remember correctly), where mining for precious metals is/was a very good business. The process that produced the veins of precious metals started with an asteroid impact melting the rocks, while the differences in temperatures at which each substance solidifies, and the speed of the process, made the precious metals flow into fractures, also produced by the impact, thus producing these veins. Pretty well demonstrated, and backed up by several lines of evidence. Sounds complex, but nothing unnatural about it. But maybe I digress ...

  5. N.E.,

    I just don't see the equivalence of physical laws and computer program rules. The patterns in nature you describe that come about from natural processes, like say, ripples in the sand, these patterns can be anticipated by knowing all the physical parameters involved. I believe you can say that the patterns are entailed by the physical processes/laws. But not so with evolutionary change because this requires undirected change in a rather precise genetic code. Mutations and changing environments cannot be predicted or expected. The physical processes that provide the new variation are separated if you will from the meaning of the change itself. The sunlight that may cause a genetic mutation has no ability in itself to cause an expected change in a DNA sequence. The natural patterns you describe also entail very little information. They are complex, but not specific. In the ripples in the sand above, an informational analogy would be to see further down the beach, JOHN LOVES MARY written in the sand. This is easily distinguishable from ripples in the sand. Physical processes by themselves cannot produce this level of specificity with the complexity. That intelligence is required is a scientifically sound conclusion even if the author is unknown and the truth of the statement cannot be discerned. You know it is language. It communicates something that is not inherent in the sand, wind and waves. Whereas the ripples are completely understandable by the sand, wind, and waves.

    1. Well Ray, I think, but might be wrong, that the reason you don't see the equivalences is because you don't want to. What I am saying is that evolution has not happened in a vacuum, and that thus, what we see cannot be purely random atoms moving around. We see only what's possible given the way nature works, and thus, disputing the power of genetic algorithms because they contain rules would be nonsensical. They show that given some basic rules, "random mutation," selection, reproduction and recombination can give us quite complicated stuff. Evolution is not writing "John loves Mary" in the sand, but selecting for those ripples that solve the problem of survival, while rejecting those ripples that cause trouble. All the while allowing ripples that have no effect in fitness. (Not all ripples are equal, right? A stone here, some more of wind there. Less sand to begin with there, more here.) Example, you seem to think that a protein has to have a very specific sequence, but far from it, they show enormous plasticity. There are so many synonyms that they might as well look, to the outsider, as any other ripples in the sand (this is what makes it hard to distinguish, say, protein-coding genes in our genome from noncoding regions, and it is worse if we wanted to distinguish regulatory regions from uncompromised ones).

      Anyway, I don't think there is any further progress to be made in our conversation. So I leave it here.

      Thanks for a conversation free of creationist propaganda. This is refreshing.

    2. N.E.,

      Likewise! I agree that we have gone as far as we can go. Besides, according to Jerry Coyne, your decisions and my decisions are determined by prior genetic history and unique environmental experiences so they are materialisticcally determined. Neither of us can be "blamed" for our conclusions. :-)