Archive for the ‘Orson Scott Card’ Category

Science Fiction Round 27: Evolution Does Not Work That Way

There is something that we’ve touched on in several previous posts, but I’d like to talk about in a bit more detail: how biological evolution as a process and evolutionary biology as a model of it are misrepresented in media in general and in science fiction in particular.  Again, almost all of what I’ve written here is not original.  And as you’ll see, the particular misunderstandings I’m discussing here are based on two problems: many people not understanding evolutionary biology, and the human tendency to assume intent and direction even in entirely random processes.

Evolution Works Only On Populations

Evolution.  You keep using that word.  It does not mean what you think it means.

Evolution. You keep using that word. It does not mean what you think it means.

No, individual organisms do not evolve.  Nor are acquired characteristics passed from creatures to their offspring.  And nor does evolution allow violations of the laws of physics.  Got it?

Evolution Has No Goals

It should be obvious, but evolution is not a conscious entity.  Evolution is simply the set of processes that drive changes in the frequency of inherited characteristics in populations over time.  This means that evolution has no goals and no design.  Crucially, humans in particular and intelligence in general are not intended outcomes of a system in which evolution can occur.  I’m thinking of Star Trek and the new Battlestar Galactica in particular here, but the same mistake is made in a lot of science fiction.

Likewise, while symbiotic and cooperative behavior can certainly evolve, nothing like the entire-biosphere-is-one-organism of Avatar or various other misunderstandings of the Gaia hypothesis can actually happen without comprehensive engineering – a system like that only works when imposed top-down on the whole biosphere, and since evolution has no goals all that can occur are bottom-up emergent properties.

Evolution Has No Direction

Since there is no goal to evolutionary processes, it isn’t the orderly progression from pond scum to bilateria to basal amniote to mammalia to simian to ape to human to glowing space being seen in 2001 and Babylon 5 and on far too many t-shirts.  Species aren’t “more evolved” or “less evolved” than one another.  All things currently alive on Earth (and a few things that are alive off of Earth) come from different lineages that have survived for the same amount of time (3.8 billion years, give or take).  Over 99.9% of species that have ever existed on Earth are now extinct – or rather the distribution of traits in any living descendants they had has changed to the point that interbreeding and producing fertile offspring with their ancestors would be impossible.  Either way, the current set of living organisms isn’t anything inherently special in evolutionary terms; we’re simply the current state of a constantly-changing and boundly-chaotic system.

Massively simplified phylogenetic tree, from the Interactive Tree of Life website ( ).  We all are buried in the "Animalia" line.  Note how similar we and all other animals are genetically and how different the microbes can be.

Massively simplified phylogenetic tree, from the Interactive Tree of Life website ( ). We all are buried in the “Animalia” line. Note how similar we and all other animals are and how different the microbes can be.  Genetically speaking, you and me and a slime mold have more in common than the slime mold does with blue-green algae.

And even the messiness of the phylogenetic tree above and the full version that it is drawn from only begins to capture the chaotic complexities of evolutionary processes.  For example, it takes no account of populations with incomplete divergence, where the definition of “species” gets fuzzy because creatures can still have offspring with one another – be those critters flowers or gulls forming ring species populations; bears that were thought to be several different species before genetic sequencing came in; or the various different clades of H. sapiens – the main H. sapiens sapiens plus neanderthalensis, Denisova, and possibly others that we all are descended from.

But the data this chart is based on illustrate something very important.  The chart is based on the genomes of the organisms concerned; charting the statistical similarity between the sequences of their component genes.  Those genes have changed in many different ways after the ancestral population of each pair of lineages split.  Sometimes, genetic information is copied.  Sometimes, it is added to.  And sometimes it is deleted.  Fundamentally, many of those alterations are random: a cosmic ray hit scatters a base pair; a DNA polymerase makes an error that is not corrected by a proofreader; and so on.  That makes them impossible to predict.  While you can put limits on the possible traits that could appear and spread in a population, there is no way to predict exactly what a population’s descendants will evolve into in the future (and if we were able to run history twice from the same starting point, we’d get two very different outcomes).  Nor can you say exactly what a population’s ancestors were like.  At best, you could do a statistical best-guess reconstruction.   There is no latent amphibian in your genome.

Evolution Is More Than Natural Selection

This one is really annoying, because it tends to show up among people who nominally have learned something of evolutionary biology.  Basically, it is the tendency to assume that every observed trait of every population must be the direct product of natural selection.

Certainly natural selection is important, even over relatively short timescales.  The spread of lactase-persistence genes in the human population over the last several thousand years is a classic example.  Even in populations where lactase production does not usually continue into adulthood, a few random people with mutations that keep it active are born.  In cultures where animals that can be used to produce dairy products have been domesticated, being able to eat those products gives a big advantage in terms of calorie and protein intake.  And so the lactase-persistence mutations that happened to be around when and where humans started keeping herds have spread throughout the population (although even now, most people worldwide don’t have lactase-persistence and many people who do continue to produce lactase in adulthood don’t produce it at the same level as in childhood).

But there are many common traits that haven’t become so due to natural selection.  Genetic drift means that an allelle or trait that is neutral or even somewhat negative in terms of whatever metric of fitness is being used can still spread throughout the population, simply by random fluctuations and sampling.  This is particularly relevant during population bottlenecks and when small populations split off from larger ones.  Among many other examples, the latter is why American Sign Language exists: there was a high incidence of hereditary deafness among the population of Martha’s Vineyard, primarily among the descendants of a few deaf immigrants in the late 17th and early 18th century.

Massively simplified model illustrating the phenomenon of genetic drift.

Massively simplified model illustrating the phenomenon of genetic drift.  Colors indicate different alleles, all of which have exactly the same fitness: each model organism has an equal chance of passing its allele to the next generation.  Initially, all organisms in the population have the same allele.  Occasional random mutations occur, one of which eventually dominates the population by random chance.  I ran the model forward 21 generations, at which point the new allele was universally fixed, but there is nothing special about that point.  In another 20 generations or so, it would most likely be replaced by something else.

Genetic hitchhiking means that alleles associated with a gene that is being selected for can increase in frequency even when they themselves provide no benefit.  And “associated” can simply mean “nearby on the same chromosome” – for example, sections of human chromosome 5 may have been selected for, but hitchhiking caused a nearby part of the chromosome carrying an allele that gives a genetic predisposition to Crohn’s disease to spread at the same time.

And exaptation means that an allele originally serving one purpose may be selected for or against for some other reason – some dinosaurs with unusual scales for better thermoregulation and showing off to one another found that they were useful when jumping, and we got flying birds.

This all means that we and most other lifeforms on Earth are carrying around a lot of baggage that no longer serves or never served any useful purpose – be that an unused appendix or endogenous retroviruses – because there is not any sufficient cost to their presence and/or sufficient benefit to their removal.  Bottom line: many things about humans and other life on Earth, and about any alien life that we might encounter, will be the legacy of random events and we should not construct just-so stories just because we are uncomfortable with being the outcomes of a chaotic system.

Evolution Is Not Fate

Related to much of above:

It’s particularly grating and harmful when “evolution made it so” or “it’s natural” is used to in an attempt to justify cultural innovations or in an attempt to defend harmful behavior by individual people and characters.  I could use Star Trek references again here, but there are far too many other examples in fiction – fantasy as well as scifi.

Here’s one: Orson Scott Card’s Ender’s Game setting fails all over the place because he creates characters that are defined almost entirely in terms of psychology by their supposed evolutionary history (although Card personally ambiguously endorses intelligent design).  The Hive-Queens are unable to recognize that individual humans are intelligent even after observing our behavior covertly for an extended period of time before invading – they only get the hint after they’ve gotten into a massive fight with humans.  That makes no sense.  Evolution is not fate, and anyone capable enough to be traveling over interstellar distances and establishing colonies on dozens of planets with their own local biospheres will be able to parse “not all intelligences are exactly like me”.  Card also extends his own personal bigotry to his characters, and this as a device to endorse his own particular and dangerously wrong ideas of sexual ethics, falsely claiming as “biological imperatives” things that are entirely the inventions of certain groups of humans.

Nor are such actively-harmful misunderstandings of evolution restricted to fiction.  Evolutionary psychology does contain some valid and important insights – such as where some cognitive biases come from – but it is also invoked in misguided attempts to excuse both harmful individual behavior and elements of current culture that are both recent in time and restricted in space.  One particularly absurd example, although it’s very far from the most harmful:  The claim that the current US-centered gendering of the color pink as feminine is due to Paleolithic foraging habits and therefore justified.  In fact, that’s only an invention of the last 60 years or so – and in the US, it was preceded by a few decades where pink (along with red) was seen as masculine.

I hope the problem here is abundantly clear.  “Evolution says so!” is being wrongly used to hold aspects of culture / behaviors above question.  First, evolution doesn’t say anythingno goals and no direction, remember?  And, second, evolution is not fate.  We not irrevocably bound to any particular culture or set of behaviors out of the incredibly wide range that we can choose from.  If we can change things to make the world a better place, we should do so – regardless of how the frequencies of different alleles changed among our ancestors.   It is dishonest and harmful to claim otherwise.