I like to watch PBS; I really do. I especially enjoy their nature shows. But I have noticed recently that there’s a theme that gets repeated often on these shows and it’s starting to get on my nerves. It’s the idea that the evolution of a species’ genes is responsible for all of its appearance and behavior.
PBS isn’t alone. Most articles that you read in newsmagazines, or columns in the newspaper, will join in. It’s a belief so universal by this time that it’s not even questioned, but it should be. What’s most amusing is the way that “Evolution”, as a Being, is deified and spoken of as though it was a person. Phrases like, “…in exchange for its small size, Evolution has given this animal a powerful survival mechanism…”, or even amazing statements like, “Evolution has designed this animal to thrive in environments with almost no water,” and the like.
What’s silly, of course, is that the whole idea of evolution is that it’s supposed to happen through chance, and through the mechanism of natural selection. Impersonally. The nature shows deify Evolution, I think, because there’s just no way they can explain what they’re documenting using only chance, genetics, and natural selection. There’s a significant explanatory gap.
Here’s the crux of the problem: we can see evidence of gradual change in response to a changing environment, but the requirements of genetic change through natural selection are just too strong. So the nature shows (and all those newsmagazine articles) lapse into writing “just-so stories” with a good deal of hand-waving.
In this rant, I would like to propose some other, equally credible, agents of gradual change in a population of organisms, and discuss the importance of “satisfactory explanations” and the province of science vs. metaphysics.
The word “science” usually means, in the context of the popular media, what is done, said, and guessed by people holding degrees in the sciences, or doing work at science labs. But we are all scientists, and we all do science, just as part of being rational creatures. I heard a good definition of science way back in high school, which went something like:
As you probably already realize, a lot that goes by the name of “science” doesn’t follow these four simple steps, especially when it’s politically or theologically charged. Questions like “is there a gene for criminal behavior,” “what is the origin of life,” or “can nuclear fusion be performed at room temperature,” often have such stakes riding on the answer that it may distort the science being done.
A good example is that of cold fusion, where step 3 above, the careful performance of the experiment, was botched. Fortunately, the scientific community rushed to repeat and verify the experiment, and was unable to, denying the hypothesis. That’s an example of when the checks and balances of the science establishment worked as they should.
But there are a number of inquires for which experiments cannot be done, or the experiments done cannot be repeated or verified. What happens during these inquires is not completely science. In the case of the evolution of life on earth, the question is a historical one, so there is no way to repeat the experiment. (Moreover, the enormous time scale required for natural selection means that nobody has ever been able to, or expects to, record a start-to-finish experiment on any given species.) What ends up passing for the science of evolution, in many cases, is simply step 1, the formation of the hypothesis.
This doesn’t mean that we can’t reason about it, or make some kind of progress in our understanding of it. It just means that the kind of mental work we are doing is more philosophical, and sometimes, theological. It’s possible to do good, useful philosophy, but that kind of work shouldn’t be called science.
It is possible, then, to examine some of the claims of evolution philosophically, to ask ourselves if the processes being proposed are credible candidates for agents of the phenomena we observe. And it is also possible to use the truly scientific results of related phenomena to evaluate the claims being made. But such evaluations are more like thought experiments and not scientific experiments.
Having made this disclaimer, let’s proceed with the problem at hand: why genetic change through natural selection is not a credible candidate, alone, for all the kinds of evolutionary change we see happening in our world.
For those not familiar with Darwin’s theory (and who can you possibly be?), it goes something like this:
(Disclaimer: I have read Darwin’s The Descent of Man, but have not read his Origin of the Species.)
Darwin conceived of his theory based in part on his observation of different species of finches in the Galapagos islands, noting that each species of finch seemed uniquely fit to its environment. This observation is one that we can all make, once sensitive to it, and people continue to do so, especially in the aforementioned nature shows.
The term “natural selection” is important. There are other kinds of selection, such as artificial selection and sexual selection. “Natural selection” means that nature alone is doing the selection—that it is simply “what happens” when no other influence except environment is present. Artificial selection is the way that horses, dogs, and cats are bred. Sexual selection refers to traits that end up being reinforced in a species because they are sexually attractive to members of that species and therefore encourage breeding.
But it’s important to note the limits of selection. As a simple example, note the degree to which breeds of dogs can vary physically, as compared to breeds of cats. There is not as much room for variation in a cat’s genes. Compare the differences in a Manx cat vs. an Angora cat; then look at a Great Dane, a Lhasa Apso, and a Chihuahua.
So selection means exactly that—a selection of the traits already present in the species’ genome. Mutation is required for the introduction of new traits which are not already part of the genome.
Mutation is difficult because the mutation must (1) not be fatal to the organism, (2) not cause it to be killed or ostracized by the group, (3) help it thrive better in its environment, (4) not only thrive, but breed more than its competitors. The answer to this difficulty is usually given as time. That is, given four billion years, anything can happen. This is a postulate which must be accepted as an article of faith. There is no practical way to reproduce a four billion year experiment.
One of the examples of natural selection that I remember from my high school biology class was the lengthening of giraffes’ necks. The argument is that giraffes used to have short necks, but during times when food was scarce, giraffes with slightly longer necks could get additional food from taller bushes and trees, therefore surviving (or thriving better) and getting to mate (or mate more often), passing on their longer-neck genes.
The unanswered question is, how does the gene for neck length work? What did the proto-giraffe population look like: some giraffes with necks as long as modern ones, interspersed with short-necked giraffes? That would be required if the gene for neck length works like a toggle switch that is set to either “long” or “short”. But I haven’t ever heard anyone defend that scenario. Rather, the proto-giraffes are considered to all have had shorter necks than modern giraffes.
The question is, once the short-necked proto-giraffes are weeded out, how does the neck length of the surviving giraffes continue to increase? Is is possible for a gene to have a mutation that results in its being interpreted as “just a little longer”, without destroying the interpretation of the gene entirely? Or, is there some kind of “genetic arithmetic” that causes two long-necked parents to give birth to a child whose neck is just a little longer than either of its parents? I haven’t heard any defense of this idea either.
That is too bad, because the idea of “genetic arithmetic”, where genes represent smoothly valued functions, makes natural selection much more credible. It is then easy to imagine a population of organisms responding gradually to gradual changes in their environment. But this is not the way I have heard it explained.
The removal of existing traits is another one that is hard to explain with natural selection. Basically, in order for a trait to be completely removed from a population’s gene pool, the gene must first be present in only a subset of the population, and second, impair that organism’s breeding chances so badly relative to its peers that it can’t mate, or mates so infrequently relative to peers not containing the gene that the gene gets “crowded out” over several generations.
One often hears talk of evolution slowly removing non-essential organs and features, such as (in our case) appendixes. Considering this process within the context of natural selection, how and why would this ever happen? Why, for example, would a primate that used to have a tail have it evolved away? I cannot think of a credible way that natural selection would prune such a trait. Why would it go away if it isn’t hurting anything? What I’m talking about here are “benign” or, at worst, mildly inefficient traits.
I did read a Discover magazine article once that took on the explanation of this phenomenon. It was a good thought provoker. The argument was that during an embryo’s development, if critical proteins and tissues were “spent” on an increased brain size, instead of a tail, this results in a more fit organism. In other words, there are no “benign” anatomical features because they have a significant cost, especially during embryonic development. The idea is that there is a kind of “cellular budget” during this stage of the organism’s life, and small choices made here have an impact on its fitness to the environment, its breeding chances, etc. It’s a good argument, if it’s true that the embryonic development always results in a “balanced budget”. Is it necessarily true that the resources not spent on a less-useful tail will be spent somewhere else, and that this somewhere else is beneficial, such as increased brain size? What governs the “cellular budget”? Another gene? Then we’re back to the previous arguments. Also, consider this same argument from a positive standpoint: why does a larger brain necessarily mean smaller organs elsewhere? Why would it be a zero-sum game?
The article focused on the star-nosed mole, which lives entirely underground and whose eyes are atrophied, even covered by skin. They are truly vestigial organs. Just as with a vestigial tail, why and how would the processes of natural selection “downgrade” the eyes? As part of the “cellular budget” argument, the article pointed out that a star-nosed mole’s parietal lobe is much larger and more developed than its occipital lobe. The parietal lobe is related to hearing. In other words, the star-nosed mole, living underground, has much more use for a sense of hearing than a sense of sight, so the brain’s “budget” is spent differently.
However, this could certainly fall under the “use it or lose it” principle that a living brain uses. Remember that our own (and many other mammals’) tremendously large occipital cortex wires itself based on inputs from its own environment starting from birth. A recent experiment, which I saw on a “Nova” episode, involved blindfolding a seeing adult for one week. During this time they practiced reading Braille, while their brain activity was monitored. Then the blindfold was removed, and a couple of days later, their brain activity monitored again while reading Braille. What the experiment showed was that during the week of blindness, the brain rapidly reallocated much of the occipital cortex to the processing of Braille, from tactile input. Once the blindfold was removed, that portion of the brain was reallocated back to sight, and the subject’s Braille skills faltered, as only the much smaller sensory/motor cortex was now involved. What the experiment demonstrated was that brain resource allocation can change dramatically over an organism’s lifetime; it is not always a genetically foregone conclusion.
One important fact that the popular media (including many tooth-grating Star Trek episodes) has forgotten is that genes are only one contributor to an organism’s final phenotype, its observable appearance and behavior. Let’s look at some of the other contributors and why their effects can cause them to be mistaken for genetic changes.
During the transition from zygote to embryo, and then from embryo to fetus, an organism’s phenotype is very malleable. It is here that the organism’s genes do most of their work. But other factors can also be at work here. We are all familiar with birth defects and other deformities; these are changes in the phenotype that can be caused not only by changes in the genome but also by changes in the prenatal environment. It is also possible for there to exist “birth affects”, not just defects. In other words, there may be either benign or even helpful aspects of the prenatal environment that have their effect so early that they are mistaken for the work of genes. How healthy the mother is, her diet, and other parts of her environment (pollutants, toxins, electric fields) can all affect the way the zygote develops. If every mother in a population of organisms is exposed to the same environment, and any of that environment has an effect, good or bad, on prenatal development, you will observe a change in the phenotype of the population as a whole, without any change to the population’s genome.
There are also traits that are hereditary, but not genetic. The case of the mice with audiogenic seizures is one well-known example. Sensitivity to loud sounds, and the severity of the reaction, turns out to be passed somehow from the mother to the child in the womb, but not in the genes. This was proven by selectively breeding mice with increasing sensitivity to noise, and then using in-vitro fertilization to have mothers without the sensitivity give birth to children from zygotes containing genetic material from noise-sensitive parents. The mice born this way did not have the audiogenic seizures. The noise sensitivity was transmitted somehow during embryonic development, not at conception.
Of course, as far as Darwin was concerned, a trait was a trait, whether genetic or hereditary. But what is important to realize is that non-genetic heredity is a mechanism by which external changes in the environment can directly affect the phenotype of the next generation without altering its genome. That distinction does matter to today’s public debate about genetic determinism.
Once the organism is living on its own, its environment continues to affect its phenotype. The organism continues to adapt to fit the environment it is living in, if it can, without any change in its genome. This is especially true early in the organism’s life.
For example, kittens raised in an environment of nothing but vertical stripes do not develop the ability to perceive horizontal shapes or motion. Moreover, they never regain the ability. The ability, though latent, was not exercised, and therefore did not develop. The lack of horizontal shapes is an extreme and unnatural case, only possible in a laboratory. But if an entire population of organisms is in an environment that lacks something (like predators) to exercise some latent ability (such as how to flee and hide from predators), none of the organisms is likely to develop the ability, and the population will be perceived to have changed as a whole.
Atrophy is a mechanism that is well-observed and well-understood in a living organism. If you don’t work out, your muscles atrophy; if you don’t walk barefoot, your feet don’t develop protective calluses; if you don’t use your memory, its capacity decreases. Atrophy can happen to a single organism and it can happen to every organism in a population, for several generations. But this means nothing regarding the species’ genome.
In other words, we assume that a change in the environment forces a change in genes, but ignore the possibility that the change in environment simply exerts a uniform pressure on every organism born into, and living in, that environment.
For example, Americans’ bottoms have been observed (via measurements of stadium seating) to have increased in width over the last few decades. Nobody seriously believes that some sort of genetic change in the gene for bottom width has occurred. Rather, a change in the environment of most Americans (increasingly sedentary lifestyles) has exerted a uniform pressure on the entire population. The same goes for the tiny doors and low ceilings of pioneer dwellings, as well as other buildings that are a hundred years old or more.
Genetic change and mutation are not, by themselves, satisfactory explanations for the variety of observed evolutionary change. Other factors, including non-genetic heredity, uniform environmental pressures on a population, and individual evolution over an organism’s lifetime, are additional significant contributors to a population’s phenotype. The questions of why vestigial features disappear, and what drives incremental change, are also unsatisfactorily explained solely by natural selection and mutation.
A “purely scientific” explanation of evolution may not be either possible or rational: the questions which evolution raises enter the realm of metaphysics. This does not mean that the questions are unanswerable, but that they must be answered outside of the hypothesis-experiment-conclusion process. An example: in Robert Pirsig’s books, Zen and the Art of Motorcycle Maintenance, and Lila, he develops a metaphysics that posits Quality as being a fundamental force in the universe, creating both subjects and objects. A consequence of this metaphysics, as Pirsig applies it to evolution, is that everything is drawn toward “higher quality”, so that the inorganic is superseded by the biological, the biological by the social, and the social by the intellectual. This explanation is obviously neither provable nor unprovable by the scientific method, and must be judged in a different way, by how well it explains the most evidence. It does, at least, close the explanatory gap.
For theistic evolutionists, those who believe that a God exists who created the universe, using evolution as a tool, there is no explanatory gap: when there seems to be no way for a given transformation to have happened on its own, the explanation is simply that God nudged it along. This explanation probably seems to some to be unbearably convenient, but it should be compared to the response of an strict materialist presented with the same evidence, whose explanation is simply that science has not yet discovered how. In both cases there is an original commitment to an underlying metaphysics.
The problem facing the strict materialist, that of entirely removing a Prime Mover from their metaphysics, is comparable to the problem of explaining Mind entirely in terms of the brain. Pirsig compares the presence of Mind on the brain to that of software on hardware: the hardware does not give rise to the software, but rather makes the software possible; the software exists in the hardware and yet, paradoxically, is not comprised of the the hardware. Attempting to describe Mind in terms of brain is like attempting to describe MacOS 9 in terms of the Motorola PowerPC chip set. No such explanation can ever be satisfactory. Even if you are able to isolate parts of the software as resident in different portions of physical memory, or to identify, by watching the hardware, when certain software operations are occurring, you will never understand the software, except perhaps to the point of knowing how to damage it by damaging the hardware. In a similar way, a strict materialist prevents himself from considering any motive force except chance, and is thus consigned to an inherently incomplete explanation.
I remember reading Richard Dawkins’ book The Blind Watchmaker with great interest. The book is an answer to the argument that says, “if you found a watch lying on the ground, you would understand at once that it was designed.” Dawkins says that he, too, is cognizant of the wonder, beauty, and diversity of the world around him, but that this is not inconsistent with Darwinism. As an example of how random mutation and natural selection can give rise to the rich panoply of life around us, Dawkins shows the results of a computer program he wrote which simulated organisms with simulated genes. The genes mutated, and the organisms mated. The organisms started out with random genomes, and Dawkins selected the half of the organisms that least looked like insects, eliminated them, and reiterated. After a number of “generations”, all of the virtual organisms looked a great deal like insects. The demonstration is a powerful illustration of the point that randomness does not mean disruptive chaos, and that simple selection of random traits can produce highly organized results. The principal problem with the demonstration, from a materialist standpoint, is that the selection is not “natural”: Dawkins already knows what an insect looks like, and guides (one could say sculpts) the population of organisms into what he has in mind. This is a qualitative distinction. In short, the demonstration actually proves that mutation and selection can make a great artistic tool.
We are routinely told an incomplete story these days about the evolutionary process. The explanatory gap that exists in the story is overlooked. Comically, Evolution as an active designer is invoked in “just-so” stories about the origin of various species’ traits. Genetic determinism and a dominant materialist metaphysics create this tension.
One solution to genetic determinism is to simply be more open-minded toward non-genetic hypotheses about changes in a population’s phenotype. We have been looking through gene-colored glasses for so long that we have ascribed them a larger role in phenotype than they actually have.
The solution to the “just-so” stories is to reopen metaphysics as a domain of inquiry complementary, not antagonistic, to science, and recognize two things: that science is not required to answer metaphysical questions, and that the recognition of a metaphysical question to answer does not presuppose the answer itself.