Monday, August 16, 2010

Evolution as natural drift

Here's an argument against seeing purpose in evolution. 

"Evolution is a natural drift ... nor is [any] guiding force needed to explain [it]. ... Evolution is somewhat like a sculptor with wanderlust: he goes through the wood collecting a thread here, a hunk of tin there, a piece of wood here, and he combines them in a way that their structure and circumstance allow, with no other reason than that he is able to combine them. And so, as he wanders about, intricate forms are being produced; they are composed of harmoniously interconnected parts that are a product not of design but of a natural drift." [Humberto Maturana & Francisco Varela, The Tree of Knowledge (1987) p 117.]

This argument shifts explanation of biological features or behaviour pattern away from teleology (patterns are caused by the evolutionary advantage that the pattern confers) towards material cause and formal cause (patterns exist because they were available and consistent with the prevailing structural coupling).

Evolutionary psychologists like to argue sweeping propositions like "men tend to behave like this towards women because it gets them a greater number of descendents" or "women tend to behave like this towards men because it ensures greater survival and prosperity for their children", as if all normal men displayed this behaviour pattern, and all normal women displayed that one. They also like to argue that certain physical types are more "attractive" to the opposite sex. The fact is that men and women display an extraordinary diverse range of behaviour patterns, as well as physical shapes and sizes, nearly all of which are consistent with the continued existence of the individuals and (if they exist) their progeny. 

Meanwhile, the fact that sexual pairing is not restricted to those who fit the evolutionary psychologists' stereotypes could be interpreted as evidence for a deeper evolutionary purpose than the kind of purpose recognized by evolutionary biologists.


morgsterious said...

Interesting and very timely post (for me at least). I was just reading a related argument earlier today in the book "The Hidden Connections" by Fritjof Capra. In his chapter about emergence vs design, Capra argues that "... the teleological assumption that purpose is inherent in natural phenomena is a human projection, because purpose is a characteristic of reflective consiousness, which does not exist in nature at large."

The terms implicit- and explicit causes popped into my mind when I was trying to make sense of his writing. Implicit causes then being external forces that generate a certain behaviour and explicit causes being design or purpose that generates a behaviour.

I'm convinced that humans need a purpose other/higher than survival and reproduction but there are still a lot of emergent structures (such as evolution) that rule our behaviour. However, evolution should not try to make perfect individuals, but go towards making the entire species as fault tolerant as possible. This means that structures emerging from implicit causes (like survival and reproduction) need to be quite forgiving and adaptable.

When we get to set our own purposes, either as individuals or as groups, we can be much more rigid and less tolerant in the pursuit. These are the guiding lights that provide our existence with some stability.



ironick said...

Sounds a lot like the neutralist-selectionist debate: . Genetic drift clearly plays a substantial role in evolution, but so does natural selection. Or are you suggesting that ALL evolution can be explained by drift?

-- Nick

Richard Veryard said...

Thanks to Morgsterious for the quote from another "teleosceptic" (if there is such a word).

Ironick asks about the "neutralist–selectionist" debate. This is an interesting angle on the subject. I am inclined to see both drift and selection as part of the explanation (efficient cause, material cause) of evolution as a whole phenomenon, but I can't see that natural selection necessarily entails purpose (final cause).

The question "Why do horses have four legs?" can apparently be answered in several different ways. Mediaeval philosophers (or enterprise architects) would appeal to the essence of horseness, and say that four-leggedness was an essential property of the true horse. Selectionists might be able to produce computer simulations showing that if a six-legged horse mutation had ever appeared, it would be unfit and therefore couldn't have survived (efficient cause). Driftarians might argue that six-legged mammals is just one of the infinite number of evolutionary paths-not-taken (material cause). These explanations may all be simultaneously valid - so there is no imperative to choose between them.

What the teleosceptic resists is the fourth type of explanation, based on the idea that equine four-leggedness serves some higher purpose (final cause). I don't think the quotes from Maturana-Varela or Capra imply any position on the neutralist-selectionist debate.

ironick said...

Richard, I think we are on the same page now. Your original post led me to believe that you (and Maturana & Varela) rejected natural selection as an evolutionary cause (as some radical proponents of genetic drift do). It sounded like you rejected the notion that "patterns are caused by the evolutionary advantage that the pattern confers." And such a rejection seems to reject natural selection, which IS based on reproductive advantage.

But your reply to me seems to indicate that you DON'T reject "natural selection" entirely--hence we agree. But now I don't know how to make sense of original post.

I think you are confusing efficient and final cause. In evolutionary biology, the efficient cause of a particular characteristic or capability would be an explanation of the sequence of mutations that led to it, while the final cause would be an explanation of its function or purpose.

For example, the efficient cause of the male peacock's tail is the sequence of mutations over eons that created its current structure. The final cause of the tail is its role in courtship display.

Purely materialistic teleologic or final causes are perfectly compatible with evolutionary theory generally and natural selection specifically. See Teleology and science ( and The four causes in modern science (

That's why I don't understand why Maturana & Varela say that life not a product of design. It IS a product of design--a perfectly natural, materialistic design. The "designer" are the processes described by evolutionary theory, esp. natural selection.

Richard Veryard said...

Hi Nick

I agree that you can describe evolution as some kind of design process. For some people, the "designer" is a literal being, while for others the "designer" is merely the process itself - what Dawkins calls the Blind Watchmaker. For myself, I prefer to avoid that kind of languaging, because it is so prone to misunderstanding.

Maturana certainly doesn't deny the process(es) of natural selection. For Maturana, if I understand him correctly, natural selection is purely Efficient Cause - it is the mechanism that explains how evolution occurs. (On that basis, I believe we can regard the mutations that provide the raw material on which natural selection operates as Material Cause.) For Maturana, Final Causes exist only as a result of what Maturana calls languaging.

The peacock's tail is a brilliant piece of rhetoric, at several levels. For Darwin, it was the "exception that proved the rule" - at one point it seemed to contradict his theory, and then he managed to come up with an explanation that satisfied his theory after all (which is always a good rhetorical move for a scientist). For the peacock himself, it persuades peahens to have sex (which is always a good rhetorical move for a male bird). But the courtship display is surely part of the HOW - it is the mechanism that has allowed this ungainly and expensive (albeit subjectively beautiful) appendage to survive. We still don't (and perhaps cannot) have an final cause explanation WHY millions of generations of peahens have developed a fancy for these tails.

Maturana's work is extremely difficult, and you can find many conflicting interpretations of his ideas. What must be beyond dispute is that his ontology and epistemology represents a radical challenge to Aristotelian thinking.

Marc Lankhorst said...

It is important to recognize that the natural drift of evolution applies to the "system as whole" and not just to a single species. Co-evolution and sexual selection are both examples of this. To return to the peacock: we need not look for a reason "why" peahens prefer their cocks to have big tails (sorry for the wording...)It is simply a result of the co-evolution of peacocks and -hens.

A possible (though greatly simplified) explanation may be as follows: genetic variations in hens and cocks leads to hens with different preferences and cocks with different tail sizes. Cocks with bigger tails need to be comparatively "stronger" to survive (try carrying a tail like that when a predator is trying to get you), i.e., natural selection results in big-tailed cocks to possess some other traits that give them a competitive advantage.

Hens preferring big-tailed cocks hence produce male offspring that have this competitive advantage too, but also carry their mothers' genes for preferring big tails, which they in turn convey onto their female offspring (the granddaughters of the first hens). Successive generations thus may develop ever bigger tails and an ever stronger preference for them. This co-evolution has no purpose (that would be teleological thinking again), other that that the big tail disadvantage must have been compensated through some other hereditary trait (or it wouldn't have survived at all).

Richard Veryard said...

System as a whole - yes absolutely, but which system?

The Wikipedia article on sexual selection offers a range of different (and largely speculative) hypotheses about the supposed sexual advantages of large tails in peacocks or large vocabularies in human males. These include competition between males (pecking order) as well as choice between females. (I also found a highly readable essay on the subject by Caspar Hewett.) Incidentally, Darwin himself thought that maybe women were genetically predisposed to fancy men with beards, but then he would wouldn't he?

The methodological problem for this class of explanation is that it is very easy to construct hypotheses and practically impossible to disprove any of them, so we end up with a multiplicity of rival explanations.

Even the notion of "advantage" is suspect. You cannot directly observe advantage, you can only observe the circumstance of survival and make up a story about it. (From a philosophical point of view, these are no different from Kipling's Just-So stories.)

For example, we may observe that the sickle cell gene provides some protection against malaria, we may also observe that this gene is more common among people of sub-Saharan African origin, and so we tell ourselves a story about the survival advantage (or "function") conferred by this gene.

(Of course there are other stories we could tell about this gene, including its role in the history of slavery, but that involves looking at a different system.)

However, there are many genes for which we are unable to tell a similar story. Does this mean that some genes have a "function" and some don't? For Maturana, the "function" of a gene is only a metaphorical description, part of the semantics that is associated with the (external) observation of a biological system.

"Adequate behavior (interactions without disintegration) ... may appear to an observer as a semantic coupling, because he or she can ascribe functional significance or meaning to any behavior, and can describe the underlying physiology as if caused by these semantic relations."

H. Maturana, Biology of Language: The Epistemology of Reality (1978)

Marc Lankhorst said...

Agreed, notions of "function", "advantage" or "fittest" are only defined in retrospect. "Survival of the survivors" would perhaps be a more correct way to put it. However, there must have been some heritable trait that made the individuals carrying it have a higher chance of survival. And there are of course numerous ways in which these differences could occur, even sexual preferences on their own (without some added advantage I gave as an example in the previous post, which was just intended to describe one possible way this could have happened).

My main point is that the entire ecosystem evolves as a whole. Because of that, the advantage of any difference in traits is necessarily something that can only be observed in retrospect, because today's advantage may be tomorrow's disadvantage, since the environment, i.e., the rest of the ecosystem plus the "inanimate world" (think climate change or meteor impact), may evolve/develop in a different direction.

Thus, the whole notion of evolution having a purpose becomes meaningless. No predictions can be made at all about long-term survival or development of species; there is no "goal". Indeed, as you stated in your first post, it is a process of natural drift without a guiding force.