Sunday, July 1, 2012

Critique of Bejan's "Design in Nature", Part III

The left photo is the Lena Delta, rotated and mirrored; image taken from Wikipedia.  The right photo is a resin casting of the human lung, distinctly showing the bifurcation of the air passages (off-white color) along with the arteries and veins (red and blue color); photo provided to Glenny and Robertson by Dr. Ewald Weibel, Institute of Anatomy, University of Berne.

The items in the above photo are used by Bejan in the Mechanical Engineering article to illustrate the similarities of a natural flow system and an engineered system, with the river and lung being a "flow" from a single point to an area, or a volume.  However, the fact that they are both flows from a single point to an area or volume is the only similarity.  The actual differences between the Lena Delta and a human lung are so vast and profound, I am astounded that Bejan would lump them together to try to illustrate his 'constructal law'.

Here is what Bejan says about his constructal law and river basins:

The constructal law tears down the walls that have separated the disciplines of science by providing a new understanding of what it means to be alive.  Life is movement and the constant morphing of the design of this movement.  To be alive is to keep on flowing and morphing.  When a system stops flowing and morphing, it is dead.  Thus river basins configure and reconfigure themselves to persist in time.  When they stop flowing and morphing they become dry riverbeds, that is, the fossilized remains of earlier “live” flow systems.

This to me this is an unholy amalgam of Darwinism and thermodynamics, and I would very much disagree with Bejan’s definition of ‘life’, and his assertion that living things must flow and morph to remain alive.  That’s not even true for rivers.  What if a river were completely channeled by solid rock from its source in the mountains all the way to the sea, and not allowed to morph?  How does this ‘kill’ the river?  Why would it need to morph to persist?  Obviously, it wouldn’t.  The only thing that matters is whether or not there is a constant inflow of new water into the river system, but that is the requirement for there to be any persistent flow of anything.  So if you omit morphing, to say that “to be alive is to keep on flowing” is like saying “to keep on flowing is to keep on flowing.”  What’s the point of saying that? 

And what about dormancy? 2000 year old seed found in Herod’s palace was able to be germinated.  Would that comport with anyone’s definition of ‘dead’?  Just because the flow of any energy or water or chemicals has completely stopped does not mean that something is dead.  For actual living things, what matters is its ability to receive an input (moisture) and execute its programming to give an output (growth and development).

It should also be said that the analogy of the dry riverbed being the fossilized remains of a river is erroneous.  If the flowing river is that which is alive, then the riverbed, or the channels carved out of the landscape, is the work being performed by the system, it is not the system itself.  So if you must equate riverbeds with fossils, the dry riverbed would be analogous to finding tetrapod tracks, not a trilobite or nautilus shell.  

With that said, whether or not life has evolved, or to what extent it has evolved, or the extent to which it is possible for it to evolve, is the question.  The whole debate between Darwin and Intelligent design is in regards to these very questions, so to hold up ‘constructal’ law as a unifying law of design in nature is begging the question.

But what about the differences between river deltas and lungs?

River deltas form due to the deposition of sediment. The sedimentation is what causes the delta to form by either diverting the original flow or by the formation of new channels during flooding.  Over time, the delta grows larger and can eventually completely relocate, as the Mississippi delta has done several times throughout its history.  But it must be noted that the branching associated with a river delta is simply the result of the sediment losing energy, falling out of the flow and building up to the point that the river is forced to branch.  This is due to pure physicality, and is contingent on whether or not the sediment blocks the flow.  All that is required for there to be a river is water + gravity + any landscape.  And even if the river becomes completely blocked, it still ‘flows up’ until it finds a spot to flow back down again.  The river is going to flow so long as there is a constant infeed of water, with or without the sediment. 

If a delta does form, it is true that the delta morphs over time, but so what? The sediment is not part of the flow per se; it only either travels with the flow, or redirects the flow after it has been deposited.  Further, the fact that there is a flow at all is not dependant on the sediment.  The formation of the delta is simply the work performed by the flow.  The channels are carved and the sediments are picked up and deposited by the flow.  The river will flow whether it carves out channels or not, and if it does carve channels, it will flow whether the channels morph or not.  And if the flow stops (or ‘dies’), it is because the water was turned off, which has nothing to do with anything other than the water having been turned off.

Now what about lungs? 

Lung formation is highly conserved, in that the branching of both the air passages and the blood passages are the same for all individuals, at least up to a point.  This includes number of branches, length of the branches, angle of branching, etc.  I am not a doctor, so I don’t know to what degree this branching is conserved (i.e. when does the branching become unique for each person, if ever), but even if the number of clusters of alveoli is not conserved, the lung must still maintain a certain shape, for the size, shape and number of lobes are conserved as well.  River deltas are not like this, for they are formed by the random deposition of silt, hence their random branching and shifting.  Lung formation is algorithmic, following a set of rules (e.g. grow until this long, then divide in two at these angles, grow until this long, then divide in two at these angles, do until this length or volume is achieved, terminate with alveoli).  Lung growth is algorithmic, not random like the formation of river deltas.

After the lung structures have been formed, the flow of the fluids can then ensue.  But it must be noted that the fluid flow begins only after the passages (i.e. channels: bronchi, arteries, etc.), are already formed. The passage formation is independent and prior to the fluid flow through the passages or channels.  Rivers form the channels in which the rivers flow, and the river basin or delta forms subsequent to the flow of water; but because the lungs form before the flow of air and blood even begins, something else entirely other than said flow has obviously caused the growth and development of the lungs.

So with the lungs we actually have two distinct flows.  The first is a flow of materials during the process of lung development, in that the air and blood passages must be manufactured, and are built on and bifurcate from the previously built bronchi and arteries. But it is only after these passage ways have been constructed that the secondary flow of air and blood can happen at all.  The first flow of materials is controlled by a set of instructions, and the second flow of fluids is controlled and directed by the structures built by the first flow.  Rivers and river deltas are not like this at all.

In light of this, it should be painfully obvious that Bejan has completely ignored the ways in which these structures are built, which are so fundamentally disparate that I don’t understand why he conflates the two, other than in an effort to sell his constructal theory to the uninitiated.