Put the word “evolution” into Google image search and most that returns are variants of “The March of Progress.” The image, looking left to right, with the knuckle walking chimp-like ancestor evolving into the human walking upright, implicitly embodies many people’s view of evolution as a perfecting progressing process: the direction of travel, both literal and metaphorical, is clear. But, Hurst notes, if these progressive processes are all there is to evolution, how come so much of our biology—especially our genetics—seems so perverse. How come we seem less than perfect?
Can you give us some examples of aspects of our biology that you consider perverse?
Laurence D. Hurst: There are so many! About half of human fertilized eggs have the wrong number of chromosomes and die almost immediately. Short erm persistence of some in part explains our high miscarriage rate. The same problems are not seen in fish and frogs. About 5-10% of us have a “rare” disease, most of which are genetic. This high number sits at the feet of our mutation rate that is unusually high. The textbooks will tell you that DNA’s job is to make RNA which is translated to protein. But only about 1% of our DNA does this job. Not so in lowly bacteria and yeasts where the great majority codes for protein (or is associated with on/off switches to control protein production). Even at the micro-level we see some oddities. Genes code for proteins and to do so, at the end of the encoding sequence, they have a stop sign—so-called “stop codons.” There are three options, one of which is better than the others as the system is most likely to actually stop. The great majority of organisms mostly employ this best one. We both overuse and conserve the least good one.
I can see now why you have called your book The Evolution of Imperfection. But what do you mean by imperfection?
LH: When I talk about the evolution of imperfection, all I really mean is that there exist a whole serious of features of our biology (or the biology of other species) that don’t make obvious sense—the examples above are such instances. The fact that they don’t make obvious sense is what defines them as imperfections and, at least to me, renders them problems worth addressing. My thesis is simple: if there is something missing in our view of evolution it is by looking at such cases that we are most likely discover what that might be.
Is that then the idea behind the book?
LH: In part, yes. But it isn’t just about asking good questions and challenging preconceptions (although I hope I do that). More importantly, in the last 20 or so years there have been two trends that have happened in parallel but that have, outside of rather niche evolutionary circles, rather stayed under the radar. One is the flood of DNA sequence data from across the tree of life that both raises new evolutionary problems and helps solve old ones. That we have so little protein coding DNA is one such problem.
The second is the development of evolutionary mathematical models that point to processes that break the simple progressive notions of evolution.
What is especially exciting is that recent evolutionary models explain much of the odd new data coming from sequencing of DNA. For example, we now have a robust—non-progressive—model that predicts that we should have DNA bloat—most of our DNA will be rather pointless, as it seems to be. Likewise, it correctly predicts that we will have a very high mutation rate, causing a lot of genetic disease. Even in many professional biological circles these new ideas are all but unknown. Indeed, there is a common tendency to see everything through just one lens, the concept of adaptive progressive evolution. It is simply assumed that there must be a function for everything that we see. The new theories say that this simply isn’t true.
In The Evolution of Imperfection I’d like to bring this newer synthesis into circulation. When we look at our genome and our cells, we need to stop thinking that we are looking at a perfectly organised city. If our cells are cities there are fires, traffic accidents, cars running the lights, not because these things are advantageous for us, but because we just can’t stop them from happening. If you witnessed a pedestrian being hit by a car you wouldn’t assume the function of cars is to propel pedestrians. But we tend to employ just such rationale, and only this rationale, when think about cells and organisms: everything that happens, happens for a good reason we often assume.
Surely natural selection is about “survival of the fittest”?
LH: Good point! And this rather gets to the heart of the issue. The classical view of the evolutionary process is that the core process is natural selection (alias survival of the fittest). A mutation that benefits the bearer will be in more offspring and hence will be at a higher frequency in the next generation and then the generation after etc. Conversely, mutations that harm the bearer will be eliminated from populations. There is nothing wrong with this. But it isn’t the whole story.
For example, if we think about mutations that have only a small effect on fitness the logic starts to break down. The nearly neutral theory of evolution notices that when populations are small, chance changes in a mutation’s frequency can make a big difference. Mutations that harm you a little can easily increase in frequency, despite selection. Think of mutations more like particles of dust floating about in a room, less like a helium (or lead) filled balloon. This model predicts that selection becomes less efficient when populations are small. In turn we expect high mutation rates and genome bloating to be seen when populations are small, not because they are an advantage but because, although harmful, they are just not so harmful for selection to be able to always act to promote perfection.
The other problem with the simple selection ideas is that they assume that “benefit” must apply at the level of the organism: if a mutation benefits me it will be seen in the future, if it harms me it will not. But we have known for a while that this is not always true. Back in the 1950s it was understood that in principle this wasn’t always so. For example, mutations that kill sperm of the same male that don’t contain the same mutation reduce male fertility (bad for the bearer) but nonetheless spread in populations. Part of my synthesis is that having our mode of reproduction predisposes us to similar traits that can harm us but still become common. With a placental mode of transfer of resources, the game is profoundly affected by the fact that early death of an embryo saves time and energy, allowing reallocation of resources. You end of with some seriously weird results, such as mutations that become common because they kill embryos as soon as possible! We see these in mice.
So we are genetically rather impoverished, is that what you are saying?
LH: To my mind the real oddity is just how poor we are on most measures. The problem with making eggs that are doomed to die almost immediately is seen in other mammals (theory predicts it to be associated with the compensation for early embryo loss) but appears to be especially common in humans. There is a lethal disorder of pregnancy, preeclampsia, that is the third greatest killer of mothers in childbirth (after blood loss and infection) and is human-specific. Generally, mothers have a 1% chance of death in childbirth, a rate not seen in other mammals. We also now know that we have one of the highest mutation rates of any organism. Some organisms have a mutation rate three orders of magnitude lower than ours, even allowing for differences in total amount of DNA. It is very hard to see us genetically as the end point of some March of Progress.
I get it. But when you talk about imperfection, are you also talking about moral worth?
LH: No! Absolutely not. But I hear similar arguments all the time. People tell me that such and such a trait isn’t favoured by evolution (by which they mean natural selection) so must therefore be morally wrong. Leaving aside their understanding of the evolutionary, process, I really struggle to see why people make this connection – but make it they do. Arguments like these seem the moral equivalent of hardened gum that you just cannot shift from your shoes.
Consider for example killing kids (infanticide) and giving your money to charity. I would rather hope that the readers will agree the former is an abomination, the later a virtue. But evolutionary biologists and economists both see altruism, such as giving away your money, as a deep intellectual problem: if we seek to maximize fitness why help others at a cost to yourself? It should be selected against. If we connect evolutionary disadvantage to moral incorrectness, you need then to conclude that charitable giving is a moral ill. Conversely, in lions and baboons, males will invade a local population and kill the young. The resident mothers then stop looking after these kids (they are dead) and become available to newly invading males. It makes evolutionary sense as regards the males’ behaviour. Similar arguments have been raised to suggest why step-fathers are more likely to kill unrelated kids, although the data here is less clear. It shouldn’t need saying but the fact that infanticide has a logically reasonable selectionist rationale should not render the killing of babies and toddlers ethically virtuous. There is not, and should not be, any obvious connection between that which natural selection favours and moral worth.
Indeed, if you think about it much of our medicine is trying to circumvent the action of natural selection. The death of a kid because of cancer would, in evolutionary terms, be seen as the operation of selection reducing the frequency of harmful mutations. But it is a callous person who would then decide that on these grounds we should not try to save such kids. Replacing hearts, taking antibiotics, drinking clean water all keep us alive blunting natural selection’s activity but are a moral good.
This is not to say that medicine doesn’t have ethical issues. As I discuss, we are on the edge of a new genetical medicine where what is technically achievable is in advance of what is agreed to be ethically permissible. We now have lots of new medical tools to mollify the impact of our rotten genetic lot, from assisted reproduction, to gene therapy, from novel drugs to prenatal diagnostics. But what should be permitted is not obvious. I’d also argue that our limited understanding of what is, and what is not evolutionarily functional, in our DNA should give pause for thought as regards modifying our DNA.
About the Author
Laurence D. Hurst is Professor of Evolutionary Genetics at the Milner Centre for Evolution at the University of Bath, UK. He is the author of The Evolution of Imperfection.