Friday, October 28, 2016

Irreducible complexity

It's been a long time since I read Behe's paradigm-examples of irreducible complexity, so I don't have a considered opinion on the cogency of his examples. Instead, I'd like to comment on this objection:

Keith Parsons 
Behe defines "irreducible complexity" as follows (from Darwin's Black Box):
“By irreducibly complex I mean a single system composed of several well-matched, interacting parts that contribute to the basic function, wherein the removal of any one of the parts causes the system to effectively cease functioning. An irreducibly complex system cannot be produced directly (that is, by by continuously improving the natural function, which continues to work by the same mechanism) by slight, successive modifications of a precursor system, because any precursor to an irreducibly complex system that is missing a part is by definition nonfunctional.”
The key argument here is this: There are complex systems such that the removal of any one of many interconnected parts renders the whole system nonfunctional. Therefore, no such system could have been built up incrementally by successive modifications of a previously functional system, but had to be created as a functioning whole.
We know, however, that, in general, this conclusion simply does not follow. Electrical grids are instances of systems that meet Behe's definition of irreducible complexity. They are complex systems constituted of interacting parts, each contributing to the function of the whole such that the removal or disabling of one part of that system can render the whole nonfunctional.
Consider the great blackout in the N.E. United States and Canada that occurred on August 14, 2003. At 4:10 PM ET, twenty one power plants shut down in three minutes, affecting fifty million people, and disrupting everything from trains and elevators to cell phones. The cause was traced to a single power plant in Ohio which shut down unexpectedly when overgrown trees contacted power lines.
Using Behe’s reasoning we would have to conclude that the power grid did not come into existence incrementally, but was produced all at once in a stupendous feat of simultaneous creation. This, of course, is absurd.
Perhaps the reply would be that power grids are obvious examples of intelligent design (design, yes, but the “intelligent” part might be questioned) since they are made by us. Intelligent agents can build up such complex systems incrementally, but nature cannot. Ah, but this is a very different kind of argument, and one much harder to support. The whole intuitive appeal of “irreducible complexity” is that it appears to confront us with a stark choice: EITHER complex biological systems are built up slowly and incrementally (i.e. by Darwinian evolution) OR they are made all at once by the activity of an intelligent Creator (God says “Let there be…”).
But to concede that “irreducibly complex” systems are not, after all, irreducible, and can indeed be built up incrementally from functional precursors, means that a much different and much heavier burden of proof is placed on the intelligent design advocate. They can no longer deploy “how possibly” arguments, since they have conceded “yes, possibly.” They can no longer point to instances of alleged irreducible complexity, such as the bacterial flagellum, and embarrass evolutionists by demanding that they show how possibly such systems could gradually develop. They have conceded that such systems could gradually develop, but now have to argue that it could not have been by natural means, and it is not at all clear how they could show this.

Let's consider a simpler illustration: I plug an appliance into the wall. The appliance works!

I move the appliance, requiring me to plug the appliance into an extension cord, which I plug into the wall. The appliance works!

I move the appliance yet again, requiring me to use a second extension cord. (The fire inspector might not approve!).

If I unplug any of the three cords, the appliance won't work. The functionality of the appliance now depends on serial connections. 

Point being: there are cases in which it's possible to incrementally construct something that's functional at each stage that's dysfunctional if you subsequently remove an element. The challenge, though, is considering analogues in nature. It took an external agent to plug the appliance into the first extension cord, then plug the first extension cord into the second extension cord, then plug the second extension cord into the electrical outlet. That's not the result of internal development, but intervention from outside the process. 


  1. Just want to add to your electrical cord illustration and an apparent hard weakness. If that intact electrical cord is so long, the electricity flowing through the wires will break down and the appliance will not work. The cord is intact. The appliance is intact. The method of applying the electricity to the appliance is not. Even the electricity to a certain degree is intact until or unless is goes beyond its ability to remain constant.

  2. I think Parsons misses the fact that an electrical grid (stripped down to just the portion he's talking about here) doesn't fit Behe's definition of being irreducibly complex. After all, Behe included right there in the part Parson quoted: "An irreducibly complex system cannot be produced directly (that is, by by continuously improving the natural function, which continues to work by the same mechanism) by slight, successive modifications of a precursor system, because any precursor to an irreducibly complex system that is missing a part is by definition nonfunctional." Thus, if you can demonstrate that an electrical grid gets improved through slight, successive modifications of a precursor system, then it's not irreducibly complex by definition.

    But I think there's another error going on here. Parsons is ignoring just how complex it is to multiply the power grid in physical space. Sure, we can *conceptualize* it as a simple thing, but it's not like the wires self-organized into a new power grid in the presence of an existing power grid or anything like that. Rather, it has to be designed, materials have to be moved, and workers have to hook it all together.

    Thus, Parsons is here confusing simplicity at one level with simplicity at all levels. To put it in perspective, all books written in English use the same 26 letters (with capital and lowercase variants). Parsons would seem to argue this means that Dan Brown's writing is just as complex as James Joyces's! Or, for that matter, that computer generated random text is just as complex as Shakespeare is.

    Clearly we can differentiate between random text and intentional text, so the fact that they're built from the same "simple symbols" says nothing about whether it is possible to turn random text into Hamlet's monologue by altering one character at a time while still maintaining meaningful text.

    And that doesn't even get into how the 26 symbols making up the alphabet even began to have the properties they do; or, to go back to Parsons, how the original power grid got started; or, to further go back to evolution, how life even began in the first place.