“Irreducible Complexity” May Be Part of the Definition of Life
There are many bad counter-arguments to Michael Behe’s famous irreducible complexity conundrum, and (in my opinion) one pretty good one.
For those unfamiliar with Behe’s argument, it goes like this:
Darwinian evolution is supposed to build complex systems gradually, overcoming vast improbabilities in tiny steps over billions of years. But, strangely, many systems in living organisms are “irreducibly complex” — they contain a core set of key elements that are all absolutely necessary for the system to function at all. Gradual evolution through random variation and natural selection could never build such a system, because the system would have no adaptive function until it was already completely finished.
After Behe made this case in his 1996 book Darwin’s Black Box, scientists (and non-scientists) scrambled to rebut him. Some argued that the systems in question weren’t really irreducibly complex; others, that they could have arisen through cooption of parts from other systems; others, that they emerged as reductions from larger complex systems that were not irreducibly complex… and so on.
None of those arguments have held up to logical or empirical scrutiny. But I don’t think those argument are the real reason that most who find Behe’s argument unpersuasive find it unpersuasive. I suspect that the real objection for most people is something more gut-level and foundational, which might be expressed something like this:
Okay, so maybe it’s hard to see how gradual, blind processes could produce a few special systems like the bacterial flagella. Because of irreducible complexity — got it. But Darwin’s theory still makes sense for everything else. So are we really going to throw out the whole theory on the basis of a few things we can’t explain? Isn’t it more likely that there’s some explanation for these things, and we just have to wait for it?1
After all, if Darwinian evolution works in theory, then it seems to follow that Darwinian evolution should have happened. And then, if living organism don’t look like they were made by Darwinian evolution, the question just becomes, “So where the heck are the things that were made by Darwinian evolution?” Even if the presence of irreducible complexity shows that all the organisms we study didn’t arise by Darwinian evolution, it doesn’t explain why they didn’t arise by Darwinian evolution.
Confusion and Mystery
In other words, for the irreducible complexity argument to persuade someone away from Darwinism, it’s not enough to show that some structures in living organisms don’t look like they were made from unguided Darwinian processes. As long as unguided Darwinian processes work in theory, the existence of irreducibly complexity in life may add confusion and mystery, but it doesn’t do away with the theory. For the argument to be really convincing, you need to also show that Darwinism doesn’t actually work to construct living organisms even in theory.
Might this be the case? Well, it would be the case if irreducible complexity is actually necessary for living systems. If something needs to be irreducibly complex in order to achieve the characteristics that would make us call it “alive,” then Darwin’s theory doesn’t even work in theory, and the mystery is solved — we see features that Darwinian evolution can’t explain simply because Darwinian evolution didn’t actually happen, and can’t happen.
Behe argued something like this in response to the criticisms of his first book. But it was at first an open question — there is no quick-and-easy way to tell if irreducible complexity is intrinsic and necessary to life, or not.
It’s extremely interesting, then, that the prominent theoretical biologist Stuart Kauffman has been promoting a definition of life that entails irreducible complexity — though Kauffman (who is unsympathetic towards ID) doesn’t use that term.
A Definition of Life
Kauffman has been arguing that what sets living organisms apart from non-living things, and what makes them able to function and to evolve, is that in living organisms the parts exist for and by means of the whole. Kauffman calls such systems “Kantian wholes” (because the idea comes from Immanuel Kant’s Critique of Judgement). A Kantian whole, to put it another way, is a self-creating system in which everything supports and depends upon everything else.
It’s easy to see how living organisms fit this definition. Your various parts can’t exist without you — you’ll never find a brain or a spleen lying around on its own (at least, not for very long). Likewise, you wouldn’t exist if you didn’t have those parts (at least, not for very long).
It’s also easy to see that such a system is by definition irreducible complex. The “whole” — by definition — encompasses all of the parts. So, if the whole is necessary for the continued existence of the parts, then all of the parts are necessary for the continued existence of the parts — which is the definition of irreducible complexity. Not all irreducibly complex systems are necessarily Kantian wholes, but Kantian wholes are necessarily irreducibly complex.
Irreducible Complexity in LUCA
Of course, someone will probably point out that this is all very interesting philosophizing, but science is about empirical evidence. And Kauffman, as a scientist, is eager to provide it. To this end, he co-authored (with the up-and-coming origin of life researcher Joana Xavier and others) a paper published in Proceedings of the Royal Society B which seemed to show that life has existed in the form of Kantian wholes as far back in evolutionary history as we can see.
Xavier et al. took a database of metabolic reactions in bacteria and archaea (the two domains of the simplest lifeforms) and looked at which reactions they had in common. They found in the intersection of bacteria and archaea a collectively autocatalytic set of 172 reactions. (“Collectively autocatalytic” means that the set of reactions is self-creating — all the catalysts of the reactions in the set are created by other reactions in the same set; e.g. A creates B, B creates C, C creates A.) From a phylogenetic perspective, this implies that the common ancestor of bacteria and archaea — and thus presumably the “last universal common ancestor” (LUCA) itself — was characterized by complex autocatalytic metabolic cycles. In a paper in the volume Evolution “On Purpose”: Teleonomy in Living Systems, Kauffman and his colleague Andrea Roli write that these findings “very strongly suggest that life arose as small-molecule collectively autocatalytic sets.”
Kauffman and his co-theorists believe that collectively auto-catalytic sets are Kantian wholes. Therefore, they argue that life has been characterized by Kantian whole-ness from the very beginning, in accordance with Kauffman’s contention that living things are Kantian wholes by their very nature. If that’s true, then — as we have seen — that means that life, by its very nature, is irreducibly complex.
What Was the Question?
If irreducible complexity really is part of the definition of life, this solves the problem raised in the response to Behe’s irreducible complexity argument.
It all comes down to What is it that we’re trying to explain? when we invoke evolution or design. Why does life need an explanation at all? What is it that makes people, cows, mushroom, pine trees, bacteria, and so forth, so very perplexing to us?
Darwin seemed to think the problem was mere complexity, or the adapted-ness of organisms to their environment. That seems plausible at first glance, but in retrospect we should have known that it isn’t the case. A pile of sand is complex — the odds of obtaining that exact same arrangement of grains of sand a second time are almost nil — but nobody thinks that the existence of piles of sand is some big mystery.
No, the thing that makes living organisms so mysterious (one of the things that makes them mysterious, anyway) is that they are irreducibly complex: they move, act, reproduce, and grow by means of an elaborate system of interconnected, interworking parts. It’s obvious (with 20-20 hindsight) that this is the real mystery in need of explanation, and it’s equally obvious that the ability of natural selection to pile up tiny, individually useful random variations in no way explains (or even attempts to explain) how such an intricate network could come to be.
So when Behe pointed to irreducible complexity, he wasn’t noticing some random, inexplicable feature of certain biological systems and using it to attack Darwin’s theory. Rather, he was putting his finger on what exactly it is about life that makes us feel it needs explaining. And that turned out to be something about which Darwin’s insights, brilliant though they were, had nothing to say.
Notes
For example, this line of thinking has got to be why evolutionary biologist Bret Weinstein feels that “if we pursue that question [a particular problem raised by ID proponents], what we’re going to find is, oh, there’s a layer of Darwinism we didn’t get and it’s going to turn out that the intelligent design folks are going to be wrong” — even though he admits that ID proponents are pointing to genuine holes in the current theory of evolution.
