Teleology in evolution

[bascule]

(ed: AARGH, all this and I posted in the wrong forum. Can someone [Mokele?] please move this to the Evolution forum, or to pseudoscience if I'm being a little too craaaazy here...)

 

There exists one pretty obvious causal relationship in the evolutionary process which I've never really heard mentioned: the rate at which the process is able to discover useful novelty is limited by its ability to keep records and share these records with other memebers of a community.

 

For starters, all life on earth is descended from a relatively complex common ancestor which includes all sorts of complex DNA transcrption and Ribosomal machinery, all of which is needed to keep an incredibly precise (but occasionally variadic) genomic record and use it to replicate and perform all behaviors necessary to stay alive. I think we can safely assume that the very first replicator from which all life descends probably had extremely simple and therefore lousy record keeping equipment and descendants which were highly variadic from the original. After a lot of stumbling around in the dark the process produced better and better record keepers until we arrive at the highly complex eqipment which resided in the common ancestor of all life presently on earth. From the highly complex nature of our shared genome we can deduce that all other descendants of the original replicator were wiped out by our common ancestor and its descendants, such an advantage did its recordkeeping prowess provide. I am in awe of "the cell".

 

In asexual reproduction any solutions which are lost through deleterious mutations must be re-discovered by subsequent beneficial mutations. Beneficial mutations discovered by your ancestral relatives and their descendants are completely useless to your descendants, because your descendants are merely copies of you.

 

Sexual reproduction looks incredibly wasteful at first; only half of the members of the species will be able to produce descendants, thus males become very expensive (in terms of resources) gene vehicles, but the advantages of sexual reproduction are obvious. Genes which are beneficial to a particular environment are selected out of a communal pool shared among a species. Thus the recordkeeping becomes tremendously better; deleterious mutations can be sorted out while beneficial ones are kept, all shared in a pool so advancements made by one individual can be shared throughout the population.

 

But then comes the issue of behaviors... using only genes useful behaviors must be immitated by all members of a community long enough for natural selection to favor the descendants who are best at the given behavior for it to be fixed in the genome (the Baldwin effect).

 

Then evolution came up with us, Homo sapiens sapiens. Behaviors which were temporarily forgotten by a community could be resurrected by reading a written account. If a particular member of a community had trouble immitating a given useful behavior, other members of the community could provide feedback as to the problems, and use abstract descriptions to explain the intricacies of the behavior. Members of a community can dole out advice as to what behaviors they find useful, and thus those who might never have learned a particular behavior because they didn't know it existed can become aware of a particular behavior by a secondhand account and thus learn which behaviors to learn rather than simply immitating every behavior they see. And thus we have the birth of humanity's great advantage over all of the other animals, meta-information, a way to keep records about the records rather than just requiring natural selection to be the ultimate recordkeeper.

 

Evolution favors strategies which made it less blind, because the more of a record it can keep, the faster it works, or that is to say, the more useful novelty the community/species accrues per unit time. Are there any who would argue that humans have substantially outpaced all other life on earth in terms of accruing useful novelty? In the end, does it really matter how that useful novelty is stored, be it in genes, memes, meme vehicles (i.e. the Internet), or what have you... it's all behavioral evolution.

 

I think it's outright obvious that the evolutionary process favors more complex behavior for a variety of reasons, the more complex your behavior the harder you are to predict, which helps both predator and prey who both try to predict each others' behavior. More complex and intelligent behavior gives you the capacity to deal with more death scenarios which would've wiped out the line of stupider individuals. It improves your ability to locate and successfully reproduce with mates (with some exceptions in our species ). Thus the process favors the capacity to sort and store more and more complex behavior patterns. Barring environmental catastrophe, wouldn't this inevitably result in the production of sentient life?

 

The recordkeeping issue was the biggest problem the system had to solve, and nowadays us humans are doing it more efficiently than ever. Now we have a collective pool of close to all information we've ever accumulated (the Internet), and we're quickly increasing the quality of its structure and the ability of members of our species to access this wealth of information. We're the best recordkeepers ever, and we're evolving faster than ever. We work on a timescale where we can actually begin to analyze the fruits of the evolutionary process and what they truly represent from a philosophical perspective, namely what this process of incrementally improved recordkeeping trying to do.

 

To me the Singularity represents a telelogical attractor, one which if you put any man-like species (in that they are conscious animal-like replicators with a fixed lifespan and a capacity to manipulate their surroundings as they see fit as well as communicate abstract information with each other) will naturally tend towards. It's something which, it seems to me, any variadic replicators unleashed on a planet will always tend towards.

 

And keep this in mind: it's an attractor. It draws a particular scenario towards a particular conclusion, but that doesn't mean that there aren't other forces at work which can prevent the conclusion from being reached. Like evolution it's just a tendancy in a particular direction (towards useful novelty) but that doesn't necessitate that any particular outcome will be reached.

 

Won't any evolutionary process which begins with a microscopic, variadic replicator always tend towards the production of a sentient species (and all subsequent evolution that entails)?

 

Anyway, I see so many trends throughout the evolutionary process which aren't ever discussed in ethological circles... perhaps the most notable being Kurzweil's Law of Accelerating Returns... in the end it's all evolution and I think there's so much yet to discover in the evolutionary process than what biologists/ethologists give it credit for, and it seems like some of the discoveries yet to be made are thoroughly rejected by the overwhelming mentality of their community, which sadly seems to have developed as a response to IDiots who have their own cockamamie telelogical ideas with which they attempt to contradict the underlying ideas of evolution... rather than attempting to extrapolate on top of them.

[Mokele]

Well, a few things:

 

First, complexity isn't favored, or even really a 'goal', but merely yet another tool. Look at bacteria. They can survie in the most hostile environments on earth, evolve at incredibly rapid rates, reproduce like crazy, and exploit practically any habitat on the planet (and several elsewhere possibly, as the survival of bacteria inadvertently launched along with various space probes over the years indicates). As a second example, look at many parasites. these animals often arose from complex, sophisticated ancestors, but have lost their complexity, trading it for sheer reproductive output. Many are little more than sacks of gonads, despite fairly close relationships to more complex ancestors. Complexity allows more options, and may be beneficial in certain situations, but the same can be said for a tail. Complexity should be viewed as an adaptation, used in appropriate environments by some organisms, but certainly not the end-all, be-all.

 

As for sex, most data indicates that not only does it not improve the evolutionary rate or sucess of lineages, but that in animals it evolved to impove genetic mixing and recombination to fight parasites. Sex-equivalents, like conjugation in bacteria, are best explained in terms of selfish-gene theory.

 

For behaviors, you seem to imply that all members of a community (or at last a substantial fraction) must employ it before it becoms selectively favored, but again, this is not always the case. If just one animals receives a mutation that improves it's hunting efficiency, that will be selected for, regardless of it' current frequency. Now, frequency-dependent selection can occur in some systems, possibly including behavior, but this does not necessarily hold true for most systems, nor prove anything teleological.

 

However, humans are where I feel it truly breaks down, because you consider meta-information, which can be consciously controlled and modified in the face of the environment, in contrast to genetic information, whose modification does not appear to be selective (there was a recent idea about 'adaptive mutation' in a bacteria, but this has been shown to be false).

 

As for the tendency for life to produce sentient species, just look at the evolutionary history of the planet. We see from the fossil record you can get a sentient species from something that looks very much like a rat in only 65 million years (possibly less if mammals 'dawdled' which they seem to have done throughout the Paleocene and Eocene), so why didn't sentient dinosaurs evolve, since they had 180 million years to do so? Or sentient octopi/squid, since cephalopods have been around for about 500 million years?

 

This is the problem with post-hoc analysis; I can just as easily point to naked mole rats and say that evolution exists to produce eusocial species.

 

The concept of the singularity as an inevitability depends solely upon human conscious will towards it, somthing that is *not* guaranteed (see the Luddites and the Amish for notable counterexamples) and fundamentally different from gene-based evolution. If I want a machine to do a task, I can make one deliberately, but there is no evidence that beneficial mutations are more likely in environments in which they would be useful or are needed. You cannot analogize a chemical process to a conscious one.

 

Evolution dismisses teleological arguements for numerous reasons, including:

1) unless they are predictive, they serve no purpose, and simply saying 'this was meant to be' does not give any predictive information

2) there is no known mechanism by which organisms can bias mutation towards particular locations or paticular outcomes.

3) mutation is a fundamentally random chemical process, which has been shown by the accumulation of mutations in genes or regions not under selection.

4) existing systems display evidence to exactly the opposite. Take, for instance, a large fruit in Australia which no longer can disperse without the extinct megafauna. If there was any teleology in nature we either would expect that organisms would not wind up in such situations at all, or that when in such situations, they would quickly develop just the right mutation to get out of it. Given that this has not happened yet and the declining numbers of the plant make it unlike to, this constitutes strong evidence against teleology in nature.

 

So it's not just a knee-jerk rejection based on ID and Lamarckism, but rather because teleological evolution doesn't fit the data we see, the data actually strongly indicates the contrary, and known mechanisms would not produce it (and we are reasonably sure that these known mechanisms are the predominant ones).

 

Mokele

[Kygron]

... teleological evolution doesn't fit the data we see, the data actually strongly indicates the contrary...

 

Perhaps it would if we took a broader perspective? If you consider, rather than an individule species, the entire "family tree", or "symbiotic group" or even ecosystem, the data may read differently.

 

That australian fruit may go extinct, but its ancestors likely still exist and would, in this case, have removed themselves from the situation.

[Mokele]

Even in the broadest perspective, teleology fails, because organisms and higher taxa persistently adapt to short-term immediate conditions and for short-term benefit. Consider mass extinctions. If evolution were teleological, wouldn't species "strive towards" states that would allow them to survive such events? Yet time and again, having survived an extinction, they radiate into the niches which were emptied.

 

Also, remember that taxa above species are fictions of the human mind; we create "families" and "orders" for our benefit, to assist us, and these terms have no inherent biological or evolutionary reality beyond the overall similarity of the species within them.

 

That australian fruit may go extinct, but its ancestors likely still exist and would, in this case, have removed themselves from the situation.

 

How is that in any way different from the more rational alternative of "the sister taxa of this plant never got to exploit this niche my simple chance (and because one of their own had already filled it), and their comparatively more secure future is simply because the dice fell against their cousin." There's no way to distinguish the two, and as such it's just philosophy (aka BS).

 

Mokele

[bascule]

First, complexity isn't favored, or even really a 'goal', but merely yet another tool. Look at bacteria.

 

I've been reading about bacteria quite extensively in The Ancestor's Tale. Pretty nuts, everything they've accomplished.

 

As for sex, most data indicates that not only does it not improve the evolutionary rate or sucess of lineages, but that in animals it evolved to impove genetic mixing and recombination to fight parasites. Sex-equivalents, like conjugation in bacteria, are best explained in terms of selfish-gene theory.

 

Wow, I really need to reread the section on sexual vs. asexual reproduction in the Ancestor's Tale and get back to you on that one.

 

As for everything else: I was really trying to draw parallels between human sociotechnological evolution and biological evolution, with our present state being substantially "closer to the attractor."

 

It's very true that species can find a niche and thus stop accruing useful novelty.

 

And yes, there's no evidence of teleology...

 

I realize evolution is fundamentally a heuristic search algorithm doing a stochastic walk. A stochastic walk will never get you anywhere in particular...

[CanadaAotS]

thanks for the link! very interesting

[Mokele]

Wow, I really need to reread the section on sexual vs. asexual reproduction in the Ancestor's Tale and get back to you on that one.

 

I'd actually recommend "Why Sex Matters" by Low, which discusses an interesting model system for this: a snail, cabable of reproducing asexually and sexually. Some populations use one method, some the other. The strongest correlation is with parasite load: lots of parasites means lots of sex.

 

Of course, the question isn't fully resolved yet, but I think that's a pretty good bit of evidence.

 

As for everything else: I was really trying to draw parallels between human sociotechnological evolution and biological evolution, with our present state being substantially "closer to the attractor."

 

Well, it's the whole analogy thing. You can get a *rough* analogy, but you can't really completely map the properties of a consciously-guided system onto an unguided system.

 

I realize evolution is fundamentally a heuristic search algorithm doing a stochastic walk. A stochastic walk will never get you anywhere in particular...

 

Except the walk isn't totally stochastic; it's constrained by various genetic, biological, and embryological factors. That's why no animal has wheels; wheels are great, but are biologically and embryologically imposible for an animal to produce.

 

Mokele

[bascule]

As promised, here's what Dawkins had to say about sexual reproduction in The Ancestor's Tale:

 

To do justice to all the theories would take a book - it has already taken several, including the seminal works I have previously mentioned by Williams and Maynard Smith, and Graham Bell's beautifully written tour de force The Masterpiece of Nature. Yet no definitive verdict has emerged. A nice book aimed at a non-specialist audience is Matt Ridley's The Red Queen. Though primarily favouring one of the theories on offer, W. D. Hamilton's theory that sex serves as an unceasing arms race against parasites, Ridley does not neglect to explain the problem itself and other answers to it. As for me, I shall swiftly recommend Ridley's book and the others before going straight to the main purpose of this tale, which is to draw attention to an under-appreciated consequence of the evolutionary invention of sex. Sex brought into existence the gene pool, made meaningful the species, and changed the whole ballgame of evolution itself.

 

Think what evolution must look like to a bdelloid rotifer. Think how different the evolutionary history of those 360 species must have been from the normal pattern of evolution. We portray sex as raising diversity and so, in a sense, it does: that is the basis of most theories of how sex overcomes its twofold cost. But, paradoxically, it has also a seemingly opposite effect. Sex normally acts as a kind of barrier to evolutionary divergence. Indeed, a special case of this was the basis of Mark Welch and Meselson's research. In a population of mice, say, any tendency to strike out in some enterprising new evolutionary direction is held in check by the swamping effect of sexual mixing. The genes of the would-be enterprising diverger are swamped into conformity by the inertial mass of the rest of the gene pool. That is why geographical isolation is so important to speciation. It takes a mountain range or a difficult sea crossing to allow a newly striking-out lineage to evolve its own way without being dragged back to the inertial norm.

 

Think how different evolution must be for the bdelloid rotifers. Far from being swamped into normalcy by the gene pool, they don't even have a gene pool. The very idea of a gene pool has no meaning if there is no sex. 'Gene pool' is a persuasive metaphor because the genes of a sexual population are being continually mixed and diffused, as if in a liquid. Bring in the time dimension, and the pool becomes a river, flowing through geological time - an image that I developed in River out of Eden. It is the binding effect of sex that provides the river with its limiting banks, channeling the species into some kind of evolutionary direction. Without sex, there would be no coherently channeled flow, but a shapeless outward diffusion: less like a river than like a smell, wafting out in all directions from some point of origin.

 

Natural selection presumably takes place among the bdelloids, but it must be a very different kind of natural selection from the one the rest of the animal kingdom is accustomed to. Where there is sexual mixing of genes, the entity that is carved into shape by natural selection is the gene pool. Good genes tend statistically to help the individual bodies in which they find themselves to survive. Bad genes tend to make them die. In seually reproducing animals, it is the deaths and reproductions of individual animals that constitute the immediate selective events, but the long-term consequence is a change in the statistical profile of genes in the gene pool. So, it is the gene pool, as I say, that is the object of the Darwinian sculptor's attention.

 

Moreover, genes are favoured for their capacity to co-operate with other genes in building bodies. That is why bodies are such harmonious engines of survival. The right way to look at this, given sex, is that genes are continually being tried out against different genetic backgrounds. In every generation, a gene is shuffled into a new team of companions, meaning the other genes with with it shares a body on any particular occasion. Genes that are habitually good companions, fitting in well with others and co-operating well with them, tend to be in winning teams - meaning successful individual bodies that pass them on to offspring. Genes that are not good co-operators tend to make the teams in which they find themselves become losing teams - meaning unsuccessful bodies that die before reproducing.

 

The proximal set of genes with which a gene has to co-operate are the ones with which it shares a body - this body. But in the long term, the set of genes with which it has to co-operate are all the genes of the gene pool, for they are the ones that it repeatedly encounters as it hops from body to body down the generations. This is why I say it is the gene pool of a species that is the entity sculpted into shape by the chisels of natural selection. Proximally, natural selection is the differential survival and reproduction of whole individuals - the individuals that the gene pool throws up as samples of what it can do. Once again, none of this could be said of the bdelloid rotifers. Nothing like the sculpting of the gene pool goes on, for there is no gene pool to sculpt. A bdelloid rotifer has just one big gene.

 

What I have just called attention to is a consequence of sex, not a theory for the benefit of sex, nor a theory of why sex arose in the first place. But if I were to attempt a theory of the benefit of sex; if I were ever to essay a serious assault on the 'essential feature of the situation that is being overlooked,' it is hereabouts that I would start. And I would listen again and again to the Rotifer's Tale. These tiny, obscure denizens of puddles and mossy moisture may hold the key to the outstanding paradox of evolution. What's wrong with asexual reproduction, if the bdelloid rotifers have run with it for so long? Or, if it's right for them, why don't the rest of us do it and save the massive twofold cost of sex?

[bascule]

Okay, let me explain where I'm coming from here...

 

I've been working on stochastic sorting/ranking algorithms which work on ontological structures (i.e. sorting structure through analysis via random walks)

 

This process simply amazes me as it seems to accrue immense order from an overabundance of chaos. Even with the underlying structure it's analyzing constantly changing, the values shift dynamically, trending towards a finalized structure which is never reached due to continuous change in the structure which is being analyzed.

 

What you seem to be saying is that the behavior of the system of life is so chaotic as to escape any kind of analysis. Shouldn't it be possible to look at trends in convergent evolution as revealing a pattern in the way the underlying process behaves. Shouldn't it be possible to analyze the probability that, say, starting from a single cellular replicator, that a population will make the leap to multicellularity in a given number of timesteps? As I read through The Ancestor's Tale Dawkins went over how time and time again we see these beautiful mathematical relationships in the way organisms self-organize (for example, in defense of my previous statement, that genetic variability remains more or less constant among species regardless of reproduction rate because high reproducers create larger populations and thus the gene pool becomes more diluted, whereas slower reproducers have smaller populations and thus it's easier for a variation to spread throughout the entire population. If you'd like I can quote the relevant section of the book.)

 

I just think as time progresses we'll continue to discover higher level interrelationships to the point that we can posit that starting from a single variadic cellular replicator we can define a set of probabilities that a given set of trends will arise in the population in a given amount of time (and be able to calculate that for any time interval)

[Mokele]

What you seem to be saying is that the behavior of the system of life is so chaotic as to escape any kind of analysis. Shouldn't it be possible to look at trends in convergent evolution as revealing a pattern in the way the underlying process behaves.

 

Not quite; what I'm saying is that analysis has failed to reveal anything. That may man we just aren't looking in the right way, but it might just mean there's nothing there to see.

 

For instance, we see lots of convergences, but why would those be evidence of anything more than similar selective pressures. Numerous geometric similaries would be more likely due to the constraints of the laws of geometry itself and how things work than any goal-directed process.

 

Mokele

[bascule]

I certainly did not want to paint evolution a goal directed process, with some sort of inevitable outcome somehow programmed into it from the start, or any unseen process somehow unnaturally influencing selection events.

 

If anything, I'm trying to describe the process of improving recordkeeping as a sort of self-catalyzing reaction. The better records that are kept, the faster a system can move to higher degrees of order.

 

I just think that occasionally environmental pressures may result in the selection of a sort of generalized solution which solves not only the environmental problem which was creating the selection pressure but a whole slew of related problems which affords a massive benefit to the offspring. (e.g. human consciousness)

[john5746]

Maybe evolution can be expressed as a wave function?

 

Anyway, intelligence on our level seems to be very rare so far. Eventually, nature will win and our species will be wiped out, IMO. It seems to pay to be small and humble when hard times hit, and I think this trend will continue. When that happens, intelligence such as ours will be a mere blip on the evolution timeline, so it may not happen again for a very, very long time.

 

So, the teleology would seem to be, the meek shall inherit the earth.

 

This seems to hold true for both genetic and technological evolution.

[bascule]

Well, I finally made it to the last chapter of Richard Dawkins' The Ancestor's Tale only to discover that the entire thing was about what I was discussing in this thread! So forgive me if I quote a few relevant sections here.

 

The American theoretical biologist Stuart Kauffman put the question well in a 1985 article:

 

One way to underline our current ignorance is to ask' date=' if evolution were to recur from the Precambrian when early eukaryotic cells had already been formed, what organisms in one or two billion years might be like. And, if the experiment were repeated myriads of times, what properties would be rare, which properties were easy for evolution to happen upon, which were hard? A central failure of our current thinking about evolution is that it has not led us to pose such questions, although the answers might in fact yield deep insight into the expected character of organisms.[/quote']

 

I especially like Kauffman's statistical proviso. He envisages not just one thought experiment but a statistical sample of thought experiments in quest of the general laws of life, as opposed to local manifestations of particular lifes.

 

[...]

 

Those biologists who could be said to take their lead from the late Stephen Jay Gould regard all evolution, including post-CAmbrian evolution, as massively contingent - lucky, unlikely to be repeated in a Kauffman rerun. Calling it 'rewinding the tape of evolution,' Gould independently evolved Kauffman's thought experiment. The chance of anything remotely resembling humans on a second rerun is widely seen as vanishingly small, and Gould voiced it persuasively in Wonderful Life. It was this orthodoxy that lead me to the cautious self-denying ordinance of my opening chapter; led me, indeed, to undertake my backwards pilgrimage, and now leads me to forsake my pilgrim companions at Canterbury and return alone. And yet . . . I have long wondered whether the hectoring orthodoxy of contingency might have gone too far. My review of Gould's Full House (reprinted in A Devil's Chaplain) defended the unpopular notion of progress in evolution: not progress towards humanity - Darwin forfend! - but progress in directions that are at least predictable enough to justify the word. As I shall argue in a moment, the cumulative build-up of complex adaptations like eyes strongly suggests a version of progress - especially when coupled in imagination with some of the wonderful products of convergent evolution.

 

Convergent evolution also inspired the Cambridge geologist Simon Conway Morris, whose provocative book Life's Solution: Inevitable Humans in a Lonely Universe presents exactly the opposite case to Gould's 'contingency'. Conway Morris means his subtitle in a sense which is not far from literal. He really thinks that a rerun of evolution would result in a second coming of man: or something extremely close to man. And, for such an unpopular thesis, he mounts a defiantly courageous case. The two witnesses the he repeatedly calls are convergance and constraint.

 

Convergence we have met again and again through this book, including in this chapter. Similar problems call forth similar solutions, not just twice or three times but, in many cases, dozens of times. I thought I was pretty extreme in my enthusiasm for convergent evolution, but I have met my match in Conway Morris, who presents a stunning array of examples, many of which I had not met before. But whereas I usually explain convergence by invoking similar selection pressures, Conway Morris adds the testimony of his second witness, constraint. The materials of life, and the process of embryonic development, allow only a limited range of solutions to a particular problem. Given any particular evolutionary starting situation, there is only a limited number of ways out of the box. So if two reruns of a Kauffman experiment encounter anything like similar selection pressures, developmental constraints will enhance the tendency to arrive at the same solution.

 

Lots of other great stuff in this chapter... maybe I'll post more if I have time.

[Mokele]

I'm unconvinced that convergence and constraint could result in the inevitability claimed, though. In part, this is due to random extinction due to totally artificial and external events, such as asteroid strikes or massive volcanic eruptions (and it has been shown that mass extinction is almost purely random, with only a slight propensity of some lineages to go extinct above others).

 

Mostly, however, convergence only really occurs when there is a shared developmental system; note that there are only gross similarities between insect and bird wings, and that the flight mechanism is vastly different (turbulent vs laminar flow).

 

While I agree that a lot of stuff would be the same if you re-wound to the Ordovician, it'd be a *lot* different if you re-wound to before any multicellular life arose, and thus before the common embryological foundations arose.

 

Mokele

[bascule]

I'm unconvinced that convergence and constraint could result in the inevitability claimed, though. In part, this is due to random extinction due to totally artificial and external events, such as asteroid strikes or massive volcanic eruptions (and it has been shown that mass extinction is almost purely random, with only a slight propensity of some lineages to go extinct above others).

 

Dawkins had this to say regarding mass extinctions:

 

(sorry, this is the summation of a section which is the summation of a book where Dawkins uses a lot of symbolism and references to other themes he's developed, so I don't know how comprehensible it will be out of context)

 

The returning host, now unabashedly sensitive to major themes in evolution, notes progress as one of them. But progress of this kind is not a uniform, inexorable trend from the start of evolution all the way to the present. Rather, to take up the initial quotation from Mark Twain on history, it rhymes. We notice an episode of progress during the course of an arms race. But that particular arms race comes to an end. Perhaps one side is driven extinct by the other. Or both sides go extinct, maybe in the course of a mass catastrophe of the kind that did for the dinosaurs. Then the whole process starts again, not from scratch, but from some discernibly earlier part of the arms race. Progress in evolution is not a single upward climb but has a rhyming trajectory more like the teeth of a saw. A sawtooth plunged deeply at the end of the Cretaceous, when the last of the dinosaurs abruptly gave way to the mammals' new and spectacular climb of progressive evolution. But there had been lots of smaller sawteeth during the long reign of dinosaurs. And since their immediate post-dinosaur rise, the mammals too have had smaller arms races followed by extinctions, followed by renewed arms races. Arms races rhyme with earlier arms races in periodic spurts of many-stepped progressive evolution.

[Mokele]

See, i disagree, because while any post-extinction world will of course have been influenced in some way by the evolution before the extinction, I'm highly uncomfortable with his use of "progress", plus I don't think it's right.

 

Firstly, what is "progress"? Becoming more fit? Becoming more complex? Becoming better at information storage? Becoming more efficient at food processing? "Progress" has no inherent biological meaning; at the least one of the prior list (or something else similar) must be used as the definition, and in that case, why not just state that as your metric? It's like "performance"; I'm testing that now, and I have to be *extremely* careful and explicit in what I'm selecting as my "performance metric" and how I define it. Does Dawkins define his "progress metric"?

 

On top of that, any of the above metrics are wrong. Fitness only increases over microevolutionary time; on a macroevolutionary timescale the fitness landscape changes due to environmental changes, so it's entirely possible for a species to become *less* fit over several million years due to environmental changes. Complexity, again, does not always increase; numerous organisms have become *more* simple over time, such as parasites. Other examples include snakes (loss of limbs) and plethodontids (lungless salamanders). Better at information storage? Hard to evaluate, since major improvements are so rare, and are difficult/impossible to infer from the fossil record. IIRC, there are examples of human societies which have lost writing, though. Food processing? The mere existence of herbivorous species which evolved from carnivorous ancestors shows this wrong, as herbivory is inherently inefficient both in processing and nutrition per gram ingested, due to the nature of the food.

 

Personally, I don't buy it, not without a defined metric of "progress" and not even for most metrics of "progress" I can think of. The above-quoted passage might as well say that the evolutionary history of the survivors influences the paths they will take, which says little to nothing about teleology in evolution.

 

Mokele

[bascule]

Personally, I don't buy it, not without a defined metric of "progress" and not even for most metrics of "progress" I can think of. The above-quoted passage might as well say that the evolutionary history of the survivors influences the paths they will take, which says little to nothing about teleology in evolution.

 

Okay, first I should mention that Dawkins bends over backwards to apologize for even saying any of this, and I'm editing all of that out. Furthermore, the first chapter of the entire book was dedicated to "The Conceit of Hindsight" and how it's seen as foolish to look at evolution as being the least bit progressive at all (at which point Dawkins mentions that he'll address the idea again at the end of the book. See you there, reader, some 600 pages down the road!)

 

But given that, here's Dawkins exhaustive definition of progress. I'll even throw in the bit about what he says regarding the Conceit of Hindsight:

 

In my opening chapter, The Conceit of Hindsight, I listened to warnings against seeking patterns, rhymes or reasons in evolution, but said that I would cautiously flirt with them. The Host's Return has provided an opportunity to sweep over the whole course of evolution in the forward direction and see what patterns we can descry. The idea that all evolution was aimed at producing Homo sapiens was certainly well rejected, and nothing we have seen on our backwards journey reinstates in. Even Conway Morris claims that only something approximately similar to our kind of animal is one of several outcomes - others being insects, for example - that we would expect to see recurring if evolution were rerun again and again.

 

Value-free and Value-laden Progress

 

What other patterns or rhymes do we discern if we look over our long pilgrimage? Is evolution progressive? There is at least one reasonable definition of progress under which I would defend it. I need to work up to this. To begin with, progress can be defined in a weak, minimalist sense with no value judgement, as the predictable continuation into the future trends from the past. The growth of a child is progressive in that whatever trends we observe in weight, height, and other measurements over one year continue in the next. There is no value judgement in this weak definition of progress. Growth of a cancer is progressive in exactly the same weak sense. So also is shrinkage of a cancer under therapy. What, then, would not be progressive in the weak sense? Random, aimless fluctuation: the tumour grows a little, shrinks somewhat, grows a lot, shrinks a bit, grows a little, shrinks a lot, and so on. A progressive trend is one in which there are no reversals; or if there are reversals, they are outnumbered and outweighed by movement in the dominant direction. In a sequence of dated fossils, progess in this value-neutral sense would mean simply that whatever anatomical trend you see as you go from early to intermediate, is continued as a trend from intermediate to late.

 

I now need to clarify the distinction between value-neutral progress and value-laden progress. Progess in the weak sense just defined is value-neutral. But most people think of progress as value-laden. The doctor, reporting shrinkage of the tumour in response to chemotherapy, announces with satisfaction, 'We are making progress.' Doctors do not, on the whole, look at an X-ray of a swelling tumour in response to chemotherapy, announces with satisfaction, 'We are making progress.' Doctors do not, on the whole, look at an X-ray of a swelling tumour with numerous secondaries and announce that the tumour is making progress, though they easily could. That would be value-laden but with negative value. 'Progressive,' in human political or social affairs, usually refers to a trend in a direction which the speaker considers desirable. We look over human history and regard the following trends as progressive: abolition of slavery; widening of the franchise; reduction of discrimination by sex or race; reduction of disease and poverty; increase in public hygiene; reduction in atmospheric pollution; increase in education. A person of certain political views might see at least some of these trends as negatively value-laden, and yearn nostalgically for the days before women had the vote or were allowed into the club dining room. But the trends are still progressive in more than just the weak, minimalist, value-neutral sense we first defined. They are progressive according to some specified value system, even if it is not a value-system you or I would share.

 

Astonishingly, it is only a hundred years since the Wright brothers first achieved powered flight in a heavier-than-air machine. The history of aviation since 1903 has been unmistakeably progressive, and at amazing speed. Only 42 years later, in 1945, Hans Guido Mutke of the Luftwaffe broke the sound barrier in a Messerschmitt jet fighter. Only 24 years further on, men walked on the moon. The fact that they no longer do so, and the fact that the only supersonic passenger service has just been discontinued, are economically-dictated temporary reversals in an overall trend that is unquestionably progressive. Aircraft are getting faster, and they are progressin in all sorts of other ways at the same time. Much of this progress does not chime with everyone's values - for example those unlucky enough to live under a flightpath. And much of these progress in aviation is driven by military needs. But nobody would deny the existence of a coherently expressible set of values, which at least some sane people might hold, according to which even fighters, bombers, and guided missles have progressively improved over the whole century since the Wrights. The same could be said of all other forms of transport, indeed other forms of technology, including, more than anything else, computers.

 

I must repeat that in calling this value-laden progress, I am not saying the values necessarily have a positive sign, either for you or for me. As I just remarked, much of the technological progress we are talking about is driven by, and contributes towards, military purposes. We could reasonably decide that the world was a better place before such inventions were made. In this sense 'progress' is value-laden with a negative sign. But it is still value-laden in an important sense, over and above my original value-free minimalist definition of progress as any trend into the future continued from the past. The development of weapons, from the stone to the spear through the longbow, the flintlock, musket, rifle, machine gun, shell, atomic bomb, through hydrogen bombs of ever increasing megatonnage, represents progress according to somebody's value system, even if not yours or mine - otherwise the research and development to produce them would not have been done.

 

Evolution exhibits progress not just in the weak, value-free sense. There are episodes of progress that are value-laden, according to at least some entirely plausible value systems. Since we are talking armaments, it is a good moment to note that the most familiar examples come out of arms races between predators and prey.

 

The first use of the term 'arms race' listed in the Oxford English Dictionary is from Hansard (written proceedings of the House of Commons) from 1936:

 

This House cannot agree to a policy which in fact seeks security in national armaments alone and intensifies the ruinous arms race between the nations, inevitably leading to war.

 

The Daily Express in 1937, under the headline 'Arms Race Worry,' naid, 'All were worried at the armament race.' It was not long before the theme found its way into the literature of evolutionary biology. Hugh Cott, in his classic Adaptive Coloration in Animals, published in 1940, deep in the Second World War, wrote:

 

Before asserting that the deceptive appearance of a grasshopper or butterfly in unnecessarily detailed, we must first ascertain what are the powers of perception and discrimination of the insects' natural enemies. Not to do so is like asserting that the armour of a battle-cruiser is too heavy, or the range of her guns too great, without inquiring into the nature and effectiveness of the enemy's armament. The fact is that in the primeval struggle of the jungle, as in the refinements of civilized warfare, we see in progress a great evolutionary armament race - whose results, for defence, are manifested in such devices as speed, alertness, armour, spiescence, burrowing habits, nocturnal habits, poisonous secretions, nauseaous taste, and procryptic, aposematic, and mimetic coloration; and for offence, in such counter-attributes as speed, surprise, ambush, allurment, visual acuity, claws, teeth, stings, poison fangs, and anticryptic and alluring coloration. Just as greater speed in the pursued has developed in relation to increased speed in the pursuer; or defensive armour in relation to aggressive weapons; so the perfection of concealing devlices has evolved in response to increased powers of perception.

 

My Oxford colleague John Krebs and I took up the whole matter of evolutionary arms races in a paper given at the Royal Society in 1979. We pointed out that the improvements to be seein in an animal arms race are improvements in equipment to survive, not generally improvements in survival itself - and for an interesting reason. In an arms race between attack and defence, there may be episodes during which one side ore the other temporarily pulls ahead. But in general, improvements on one side cancel out improvements on the other. There is even something a bit paradoxical about arms races. They are economically costly for both sides, yet there is no net benefit to either, because potential gains on one side are neutralised by gains on the other. From an economic point of view, both sides would be better off coming to an agreement to call off the arms race. As a ludicrous extreme, prey species might sacrifice a tithe of their number in exchange for secure and untroubled grazing for the rest. Neither predators nor prey need to divert valuable resources into muscles for fast running, sensory systems for detecting of enemies, vigilance and prolonged hunts that are time-wasting and stressful for both sides. Both sides would benefit if such a trades union agreement could be reached.

 

Unfortunately, Darwinian theory knows no route by which this could happen. Instead, both sides pour resources into competing with their own side to outrun the other, and individuals of both are forced into difficult economic trade-offs within their own bodily economies. If there were no predators, rabbits could devote all their economic resources, and all their valuable time, to feeding and reproducing more rabbits. Instead, they are forced to devote substantial time to looking out for predators, and substantial economic resources into building up escape equipment. In turn, this forces predators to shift the balance of their economic investment away from the central business of reproducing, and into improving their weaponry for catching prey. Arms races, in animal evolution and human technology alike, show themselves not in improved performance but in increased shifting of economic investment away from alternative aspects of life and into servicing the arms race itself.

 

Krebs and I recognised assymetries in arms races that might result in one side shifting more economic resources into the arms race than the other. One such imbalance we dubbed the 'Life/Dinner Principle.' It takes its name from the Aesop Fable in which the rabbit runs faster than the fox because the rabbit is running for his life, while the fox is only running for his dinner. There is an asymmetry in the cost of failure. In the arms race between cuckoos and hosts, every individual cuckoo can confidently look back on an unbroken line of ancestors who literally never failed to fool a foster parent. An individual of the host species, on the other hand, can look back on ancestors, many of whom never even met a cuckoo, and many of whom met one and were fooled by it. Plenty of genes for failing to detect and kill cuckoos have passed successfully down the generations of the host species. But genes that cause cuckoos to fail in fooling hosts have a much more hazardous ride down the generations. This asymmetry of risk fosters another: an asymmetry in resources devoted to the arms race as opposed to other parts of life's economy. To repeat this important point, the cost of failure is harder on the cuckoos than on the hosts. This leads to asymmetries in how the two sides set their balance between competing calls otn their time and other economic resources.

 

Arms races are deeply and inescapably progressive in a way that, for example, evolutionary accommodation to the weather is not. For an individual of any one generation, predators and parasites ust make life harder in pretty much the same way as bad weather does. But over evolutionary time there is a crucial difference. Unlike the weather, which fluxuates aimlessly, predators and parasites (and prey and hosts) are themselves evolving in a systematic direction, getting systematically worse from their victims' point of view. Unlike evolutionary tracking of ice ages and droughts, arms race trends are value-laden in the same kind of way as technological improvements in planes and weapons. Predators' eyes get sharper, though not necessarily more effective, because prey get harder to see. Running speeds increase progressively on both sides, though again the benefits are in general cancelled out by parallel improvements on the other side. Sabre teeth get sharper and longer as hides get tougher. Toxins get nastier as biochemical tricks for neturalising them improve.

 

With the passing of evolutionary time, the arms race progresses. All the features of life that a human engineer would admire as complex and elegant become more complex, more elegant, and more redolent of the illusion of design. In Climbing Mount Improbable I distinguished design from 'designoid' (pronounced design-oid, not dezzig-noid). Spectacular feats of designoid engineering, such as the eye of a buzzard, the ear of a bat, the musculo-skeletal apparatus of a cheetah or a gazelle, are all climactic products of evolutionary arms races between predators and prey. Parasite/host arms races culminate in even more finely meshed, co-adaptive designoid climaxes.

 

And now for an important point. The evolution of any complex designoid organ in an arms race must have come about in a large number of steps of progressive evolution. Such evolution qualifies as progressive by our definition because each change tends to continue the direction of its predacessors. How do we know there are many steps rather than just one or two? By elementary probability theory. The parts of a complex machine, such as a bat's ear, could be rearranged at random in a million ways before you hit another arrangement that could hear as well as the real thing. It is statistically improbable, not just in the boring sense that any particular arrangement of parts is as improbable, with hindsight, as any other. Very few permutations of atoms are precision auditory instruments. A real bat's ear is one in a million. It works. Something so statistically improbable cannot sensibly be explained as the result of a single stroke of luck. It has to be constructed by some sort of improbability-generating process, ratcheted up by what the philosopher Daniel Dennett calls a 'crane' (as opposed to a 'skyhook': the analogy is to the man-made lifting machine, not the bird). The only cranes known to science (and I would bet the only cranes there have every been, or ever will be, in the universe) are design and selection. Design explanes the efficient complexity of microphones. Natural selection explains the efficient complexity of bat ears. Ultimately, selection explains microphones and everything designed too because the designers of the microphones are themselves evolved engineers generated by natural selection. Ultimately, design cannot explain anything because there is an inevitable regression to the problem of the origin of the designer.

 

Design and natural selection are both processes of gradual, step-by-step, progressive improvement. Natural selection, at least, could not be anything else. In the case of design it may or may not be a matter of principle, but it is an observed fact. The Wright brothers did not have a blinding flash of inspiration and promptly build a Concorde or a Stealth bomber. They built a creaking, rickety crate that barely lifted off the ground and lurched into a neighbouring field. From Kitty Hawk to Cape Canaveral, every step of the way was built on its predecessors. Improvement is gradual, step by step in the same continued direction, fulfilling our definition of progressive. We could, with difficulty, conceive of a Victorian genius designing a sidewinder missile fully formed within his Zeusian, side-whiskered head. The notion defies all common sense and all history, but it does not instantly fall foul of the laws of probability in the way we would have to say of the spontaneous evolution of a flying, echolocating, modern bat.

 

A single macromutational leap from ground-dwelling ancestral shrew to flying, echolocating bat is ruled out just as safely as we can rule out luck when a conjuror successfully guesses the complete order of a shuffled pack of cards. Luck is not literally impossible in either case. But no gould scientist would advance such prodigious luck as an explanation. The card-guessing feat has to be a trick - we've all seen tricks that appear just as baffling to the uninitiated. Nature does not set out to fool us, as a conjuror does. But we can still rule out luck, and it was the genius of Darwin to rumble nature's sleight of hand. The echo-ranging bat is the result of an inching series of minor improvements, each adding cumulatively to its predecessors as it propels the evolutionary trend on in the same direction. That is progress, by definition. The argument applies to all complex biological objects that project the illusion of design and are therefore statistically improbable in a specified direction. All must have evolved progressively.

 

The returning host, now unabashedly sensitive to major themes in evolution, notes progress as one of them. But progress of this kind is not a uniform, inexorable trend from the start of evolution all the way to the present. Rather, to take up the initial quotation from Mark Twain on history, it rhymes. We notice an episode of progress during the course of an arms race. But that particular arms race comes to an end. Perhaps one side is driven extinct by the other. Or both sides go extinct, maybe in the course of a mass catastrophe of the kind that did for the dinosaurs. Then the whole process starts again, not from scratch, but from some discernibly earlier part of the arms race. Progress in evolution is not a single upward climb but has a rhyming trajectory more like the teeth of a saw. A sawtooth plunged deeply at the end of the Cretaceous, when the last of the dinosaurs abruptly gave way to the mammals' new and spectacular climb of progressive evolution. But there had been lots of smaller sawteeth during the long reign of dinosaurs. And since their immediate post-dinosaur rise, the mammals too have had smaller arms races followed by extinctions, followed by renewed arms races. Arms races rhyme with earlier arms races in periodic spurts of many-stepped progressive evolution.

 

Whew!

 

When it comes down to it, here's the score as I see it: Evolution was progressive up to a point, and then it got stuck. All those goddamn mass extinctions! They've held evolutionary progress in check for quite sometime; each time creatures had evolved to be great survivers in the day to day year to year century to century world, WHAM! A catastrophic event out of nowhere kills everything, and all that beautiful complexity is lost.

 

Well it seems that, at least in our case, evolution managed to solve that problem because unlike any other animal on earth we can plan on that kind of scale... a multigenerational scale when people can begin thinking about problems which will confound humanity centuries or millenia down the road. We can plan and prepare for a supervolcano eruption; we can blow up or divert asteroids before they hit the earth. The mass extinction problem is one that evolution has clearly been encountering repeatedly for quite some time, and we're it. We're the solution.

 

And our evolution has been remarkably progressive... I can't wait to see where it's going (Singularity)

 

(Ed: Apologies for the successive edits! Took me awhile to develop this properly...)