Monday 3 September 2012

Universal Darwinism vs "constructal theory"

All paradigm shifts result in casualties. On of the likely casualties of Universal Darwinism seems set to be the "constructal theory" developed by Adrian Bejan. Constructal theory is described in the recent popular book Design in Nature.

Universal Darwinism seems to be largely a superset of "constructal theory" - it explains all the same phenomena and many more besides. Where constructal theory seems associated with the maximum power principle, Universal Darwinism is more closely associated with the maximum entropy thermodynamics - which is similar, but better.

I can't be sure that constructal theory will go up against the wall - as Universal Darwinism eats its lunch. Perhaps constructal theory will succeed in carving out a niche for itself. However, it does look as though the writing is on the wall for it, written with Dennett's "universal acid".


Update 2013-11-13: I got around to reading and reviewing - Bejan's book, Design in Nature. The details are here.

6 comments:

  1. I am sorry to tell you that you are incorrect about the Constructal Law. It says absolutely nothing about "maximum" anything. It says nothing about minimum, optimum, end design, or destiny.

    The constructal law states this : "For a finite-size flow system to persist in time (i.e. to live) it must evolve in such a way that it provides easier access to its currents". This is the law of the design in nature phenomenon, and time direction of evolution, the evolutionary changes. It is a dynamic, changing mental viewing, not a proclamation of end design.

    You like "maximum entropy (production)", while there are many more scientists who argue in favor of "minimum entropy generation" (the latter do so with good reason, because it is their ad-hoc principle that accounts for animal evolution, and human engine/vehicle evolution). Had the animal and the truck been evolving toward maximum entropy generation, they would have gone extinct from the beginning of evolution. Both camps are happy, locally, inside their tents.

    The point is that one cannot unify science by speaking maximum this minute and minimum the next. Or minimum flow resistance today, and maximum flow resistance tomorrow, which is also spoken all over the literature.

    We have shown that all these (and many more, contradictory) ad-hoc optimality statements are covered by the constructal law. See for example this 2011 review in the peer reviewed journal Physics of Life Reviews (see Section 1):

    http://www.constructal.org/en/art/PLREV-2011-bejan_and_lorente.pdf

    Adrian Bejan
    Duke University
    3 Sept 2012




    ReplyDelete
  2. I should perhaps have stated that my use of the term "maximum power principle" was my attempt to translate one aspect of constructal theory into more conventional language. I did describe their relationship as an "association", though.

    Many do talk about "minimum entropy generation" - which is a rather obvious idea. However it isn't in conflict with maximum entropy thermodynamics - any more than the idea of a fridge decreasing its internal temperature conflicts with the idea of it increasing the temperature of the surrounding world. The idea of "minimum entropy generation" within organisms is simply the result of incomplete accounting.

    "Maximum entropy thermodynamics" won't "unify science". Nor will Universal Darwinism. However these are two huge and significant theories. Both cover the roughly same domain as constructal theory. I don't really see how they can all peacefully coexist. Constructal theory and "maximum entropy thermodynamics" are particularly uneasy bedfellows - since both are framed in thermodynamic terms. However, perhaps constructal theory can eek out some sort of independent niche. Or maybe those two theories will enter into each other's embrace. I'll wait and see.

    ReplyDelete
  3. I have just seen criticism of Construtal law of bejan in some other websites too. It disappoints me. People just are not able to digest certain things held close to their hearts. May be Prof. Bejan would see the day when all these people would tread this(HIS) path.When all his articles are peer reviewed, how come critiques could totally overlook them.

    ReplyDelete
  4. Well, I'd have to say I'm rather envious that you have Dr. Bejan commenting on your blog. I have attempted to somewhat engage Dr Bejan's theory, since I became aware of it. It is a rather famous theory, well known I believe in the literature with some 10,000 citations/refs. I wrote a critique of the constructal theory asking "Is the Constructal Theory provable or disprovable?" I am an empiricist, working with actual chemistry...I am interested in what can be shown, I have no agenda towards Universal Darwinism or against any other theory specifically. Selectively driven dissipative systems, some recent work appearing at MIT, would I believe, be along the llines of Universal Darwinism, no question. And constructal theory, I thought, would be in the minimal entropy generation side, with its maximal flow for the least energy expenditure (or entropy loss?), though Dr. Bejan seems to disagree with that label. I have attempted to have clarification of how constructal theory differs from evolution, since it seemingly invokes a principal of "evolution" within its statement. I am not convinced that maximum or minimum entropy generation is meaningful, why not "sufficient" entropy generation? So in that sense I'd agree as you said with the "incomplete accounting" it is to some extent a matter of either looking at entropy generation inside or outside the cell. I would certainly invite Dr. Bejan to comment at any time, to respond to my critique if he so desires.

    ReplyDelete
  5. My main critique is in my review of Bejan's book, "Design in Nature", which I link to at the end of this article. Basically, I claim that Bejan is reinventing the work of others who published before he did - and doing a second rate job of it. Bejan fails to cite or give credit. The result is at least as much marketing as science.

    ReplyDelete
  6. @MKK, satisficing is more wooly and vague than maximization - since you have to say how much is enough. If the stronger maximization theory is accurate, scientists should prefer it - since it makes more specific predictions with fewer assumptions.

    ReplyDelete