Essays - Summer 2005

Accidental Elegance


How chance authors the universe

By Mary Beth Saffo

June 1, 2005


In the century and a half since Charles Darwin’s Origin of Species, scientific research has conveyed one consistent message: evolution is an indisputable fact. Despite this message (or perhaps because of it), scientific investigations continue to be accompanied by tempestuous public debates on the political and religious implications of evolution. A significant but troublesome question centers on the role of chance in the history of life. On the level of ideas, the debate is whether there is an overarching purpose, destiny, and design to evolution or whether evolution proceeds in a less directed, even random, fashion. On the level of biology, the debate is whether evolution has been shaped only by adaptation through natural selection or whether chance has also had an influence. I suggest that both chance and adaptation play important evolutionary roles. This is not a radical notion: elementary college biology texts routinely note the place of each in evolutionary processes. But accepting the implications of this utterly conventional generality is strangely difficult for us. Especially when Homo sapiens enters the picture, philosophical and theological questions get mixed up with the biological, putting our fears and insecurities on full display.

A Templeton Foundation symposium at Harvard in 2003, “Life: Cosmic Accident or Cosmic Destiny?” showed the difficulties that arise when we think about evolution in the context of ourselves. The symposium’s organizers asked, “Does the cosmos have a purpose? Does life?” The best response of the evening came from the distinguished science historian Evelyn Fox Keller, who asked simply, “Why do we feel compelled to ask this question?”

Why, indeed?

One answer lies close beneath the surface of many arguments about evolution: the belief, whether acknowledged or not, that the purpose of the cosmos is us. Despite our increased scientific understanding of the universe over the last five centuries, narcissism remains a remarkably persistent component of the human psyche. As Ann Druyan wrote in Skeptical Inquirer, science has helped pierce our “infantile, dysfunctional need to be the center of the universe, the only love object of its creator.” But we still have a long way to go, scientists and nonscientists alike.

At the Harvard symposium, the British paleontologist Simon Conway Morris weighed in on the side of cosmic destiny. Echoing the arguments of his most recent book, Life’s Solution: Inevitable Humans in a Lonely Universe, Conway Morris asserted that adaptation is the prime shaper of evolutionary history and that humans, or humanlike creatures, are evolutionary “near-inevitabilities.” While many do not agree with Conway Morris, his reasoning nevertheless illustrates a tendency in us all. We are still tethered to a human-centered view, the Aristotelian “ladder of life” stuck firmly in our heads. Consider an article, “Directionality in the History of Life,” written in 2000 by two superb Harvard paleontologists, Andrew Knoll and Richard Bambach. Their thoughtful analysis identifies six stages of ecosystem evolution, beginning with the origin of life and ending with the most recent stage, the ecological impact of human intelligence and technology. But the graphic illustration of their ideas is so typical of us: early prokaryotes (microbes without nuclei, including bacteria) are down at the bottom, and intelligence, meaning us, is at its usual place at the top. Even the late Stephen Jay Gould, who did as much as anyone to raise our awareness of the limitations and even absurdities of human-centric perspectives, was not immune to the pervasive influence of hierarchical views of life. In the very last lecture of his famous history of life course, only 11 days before he died, Gould noted with embarrassment that, after 35 years of teaching, his course syllabus still ended conventionally, with human evolution as the final lecture topic. He promised to change the syllabus the next time around.

The evolution of human intelligence is a central question. Not only is knowledge of our social and biological history crucial to social, ethical, and medical progress, it is of even greater importance to the nurturing of the intertwined, deeply human qualities of curiosity and wisdom. Our need to wonder about ourselves and our place in the universe is a defining attribute of Homo sapiens. But if understanding our evolutionary history leads always to a conclusion that humans are the acme of evolutionary perfection, the ultimate point of evolution, or the purpose of the cosmos, then we are simply not thinking carefully or broadly enough about the universe, about life, about evolution, or about ourselves.

For starters, humans are far from perfect organisms. Like all animals, we are metabolically impoverished, dependent upon photosynthetic bacteria, algae, and plants to generate oxygen and edible carbon for us and on a tiny cluster of bacterial species to generate usable nitrogen from nitrogen gas in the atmosphere. Without the work of these organisms and other microbes, our lives and indeed the biosphere itself would come quickly to a halt.

And our anatomy could use a little touching up. First on my list would be a redesign of the sinus cavities; many others might vote for improved spinal architecture. Anyone with recent dental bills might well envy the ability of sharks and snails to replace their teeth continuously. Even our best assets have limitations. Our immune system, sophisticated and effective as it is, does not prevent all infectious disease, and it has a disconcerting tendency to attack our own cells in ways that can cause additional disease. Even the strongest of us has experienced some limitations in the functioning of our muscles or joints, our digestive and respiratory systems, our hearing or eyesight. Casual consideration of the average talk show, or faculty meeting, or the stupendous folly of environmental abuse, makes it clear that even our vaunted intelligence, a key to our evolutionary success, has its limitations.

In this, we humans are not alone. No organism on earth is perfectly designed, because the history of life has been shaped by many factors, including adaptation by natural selection. The possibility for adaptation to achieve perfection is limited by mechanics: only a finite range of structures can be built from living materials, and the functioning of any organism is constrained by the same physical laws, such as gravity, that affect the nonliving portions of the universe. It is also limited by genetic history: natural selection cannot create organisms from scratch; it can only edit body-plan blueprints that are presented to it, and some of these inherited blueprints can be modified more easily than others.

In addition, there is another factor shaping evolution—chance.

Happenstance, events out of the blue, circumstances beyond our control (or at least beyond our ability to predict), are woven through human experience. We can all name apparently random incidents in our lives that have had lasting implications: a close call, a tragedy, an encounter with a person who changes our life. Accidental events could have even larger- scale effects: If Richard III’s horse had not lost his shoe, would the Tudors have come to power? How would U.S. history be different had John F. Kennedy not been assassinated? Had the Palm Beach ballots been less confusing during the 2000 presidential elections, would the United States have gone to war in Iraq? Just the fact that we ask these questions shows our recognition of the potential for chance events to affect the course of history.

Aside from historical speculation, the emotional and economic aftershocks of last December’s tsunami prove that unpredictable events, such as natural disasters, can devastate societies as well as individual lives, in patterns that have nothing to do with justice or fairness and that even defy easy biological explanations. (Yes, some people were saved from the tsunami because they could run fast, swim well, or were alert to natural warning signs, but broadly the tsunami killed without regard to health, social status, or personal worthiness.)

If we acknowledge, however grudgingly, the role that happenstance plays in our everyday lives, and its potential to affect the larger flow of history, then why do we resist the role of chance in evolution? If chance events have the power to alter the lives of individuals or the direction of whole societies, over decades, why couldn’t chance have affected the unfolding of evolution in major ways over billions of years?

Human lives are an entanglement of circumstance (genetic, family history, chance events) and conscious choices and actions. Why shouldn’t evolution be similarly influenced by an interplay of chance and design?

Evolutionary change can be perceived in two ways, both of which offer evidence that adaptation is an important force in evolution. First, the evolution of a lineage can be tracked through the genetic and structural changes, or divergence, of a group of organisms from its ancestors. This divergence can be accompanied by diversification of descendants from ancestral forms, leading to an increase in the diversity of species over time; lineages can also decrease in diversity or go extinct. For many organisms, these changes can be followed directly in the fossil record: for instance, insect evolution can be traced from a few wingless ancestors in 400-million-year-old rocks to the million-odd, mostly winged insect species living today.

Evolutionary change can also be inferred from comparing genetic, structural, and chemical features among modern organisms. Common features can provide clues to shared ancestry, especially when those features occur despite differences in habitat or life style. For instance, despite striking differences in size, diet, and ways of life of humans, whales, and bats, all these organisms share several attributes—skeletal organization, warm-blooded- ness, and ways of nourishing their young, along with similarities in DNA sequences—that confirm their common mammalian heritage. Among the finches of the Galápagos Islands made famous by Darwin, species diversification can be correlated with adaptation to different environmental conditions on different islands.

The second way in which change can be traced is through convergence. This form of evolution also represents divergence from an ancestral type, but it is recognized in a different way: by similar structural and physiological similarities among genetically unrelated species, especially those in similar habitats. By illustrating the common challenges faced by organisms coping with similar ecological problems, and the common design solutions that can arise despite different genetic ancestry, convergence can offer especially striking evidence of the power of adaptation to shape evolutionary change.

For instance, several unrelated fast-swimming marine animals—squid, penguins, dolphins, and tuna—share a torpedo- like shape, which, for very good hydrodynamic reasons, minimizes drag in large organisms. Another impressive example of convergence is the structural and physiological similarities in unrelated desert-dwelling plants, including cacti, some euphorbs (spurges), and some agaves (century plants and relatives). Despite their differing ancestries, these species share a succulent habit, reduced leaves, and modifications in their photosynthetic metabolism (known as “CAM”) that allow them to maximize photosynthetic efficiency at high temperatures while conserving water. Adaptive evolution through natural selection favors genotypes of organisms whose structures and physiology allow them to reproduce more successfully than others. Natural selection does not inevitably lead to evolutionary change; sometimes the same genotypes continue to be favored over time. But continued increases in the percentages of particular genotypes over time can lead to evolutionary change. It is remarkable to consider the potential for transformation in what is, in effect, an editing process.

But does adaptation explain all evolutionary change?

Life’s Solution argues, in essence, that it does. While acknowledging genetic constraints and chance, Conway Morris sees these as minor players, less important and less interesting than adaptation, which he believes is the only significant force behind large-scale, long-term, evolutionary trends. Surveying the convergent phenomena that he finds relevant to the evolution of intelligence, Conway Morris sees intelligence as an adaptive inevitability, concluding that abstract reasoning and other markers of high intelligence could evolve only in a human or humanoid body.

That comforting vista, human destiny, once again. The very familiarity of this view should give us pause. Humans are unique; there may indeed be good biological reasons why abstract reasoning might be most likely to evolve in a primate-like animal. But, with a subject so close to our hearts, the assumptions underlying our conclusions warrant especially careful attention.

One cause for caution is the very complexity of adaptation, which makes it easy to misinterpret. We oversimplify adaptation by failing to consider where an organism is coming from: that is, its genomic and architectural inheritance and the environment in which it lives. Viewing adaptation in this fuller context shows that multiple, sometimes conflicting, pressures contribute to the design of organisms. Almost every adaptive solution has costs and tradeoffs: for instance, the CAM metabolism so useful to photosynthesis in desert-dwelling plants comes at an energetic cost to these plants compared to that of normal photosynthesis.

Another cause for caution is the human tendency is to assume that natural selection “sees” an organism the way we do, to suppose that a feature that impresses us is crucial to an organism’s survival. A specific manifestation of this tendency is the way we equate mammalian and, specifically, human features with evolutionary success. Conway Morris notes that the warm-bloodedness (“endothermy”) so characteristic of mammals is also convergently present in birds and some large, fast-swimming fish, such as tuna. Because endothermy is convergent, reasons Conway Morris, it is always adaptive, automatically enabling endothermic animals to invade new, cooler habitats. Endothermy is a vital adaptation for mammals, birds, and some other animals, but it isn’t everything. Despite endothermy, many birds and mammals (including ourselves, if deprived of clothing and shelter) cannot tolerate cold climates. And since endothermy entails the production of one’s own body heat, it is energetically costly. (That’s why some mammals hibernate during winter: to reduce their “fuel costs” when temperatures are low and food is scarce.) Even more to the point, more than 95 percent of animal species—most fish and almost all invertebrate animals—are “cold-blooded”; yet they have continued to evolve and flourish on earth for hundreds of millions of years, not just in warmer habitats, but in the coldest oceans and land habitats as well. For such animals, the absence of endothermy has been no obstacle to evolutionary success.

In considering the evolution of intelligence, it is natural to assume that intelligence is always adaptive, indeed the ultimate sign of evolutionary progress. While intelligence clearly matters to our evolutionary success, it does not follow that human-style intelligence is necessarily beneficial, or even relevant, to evolutionary success in other organisms. If it were, why do organisms such as insects, legumes, and bacteria—the evolutionary success of which is amply demonstrated by their large populations, wide diversity, global distribution, ecological importance, and long-time evolutionary persistence—differ so widely in amount or style of “brain power”?

Viewing evolution from the present moment—our moment—we see our world as the endpoint of evolution, but unless we succeed in destroying the biosphere completely, evolution isn’t over yet. Will the long-term evolutionary trends we so confidently predict from the evolutionary success stories we see today (including ourselves) bear any relation to evolutionary reality thousands or millions of years from now?
Finally, chance is also at work.

Chance permeates every level of the evolutionary process. Consider the importance of mutation. Genetic variation is the raw material for evolutionary change. In sexually reproducing organisms, some of that variation comes from the reshuffling of genes (“recombination”) that accompanies sexual reproduction. Genetic exchanges across species barriers—interspecies hybrids, transfers of genes between species (especially bacteria), and acquisition, loss, or rearrangement of genes among members of long-term symbiotic partnerships—generate even more variation. But the ultimate source of all the variation generated by genetic shuffling is mutation of DNA. The ultimate engine of evolutionary change, the ultimate arbiter of potential for change—even, as my student Jura Pintar has remarked, the relative timing of a change—is mutation. And mutation is a random process.

Although many mutations, sometimes even very small ones, can have major adaptive consequences (a single amino acid change in a single chloride- transporting protein can result in cystic fibrosis), many do not. For instance, the DNA code can be a sloppy language; in several cases, more than one DNA “word” codes for the same amino acid. Such “synonymous” mutations rarely have functional consequences. In addition, many organisms, especially eukaryotes, harbor bits of seemingly inactive DNA. These bits also mutate, but with few apparent functional consequences. Even mutations that do lead to an amino acid change in an enzyme protein might not make a difference in protein function if the new amino acid is chemically similar to the original and if its position is far from “active sites” (regions binding to other molecules, the sites usually least tolerant of structural changes). In short, some mutations are more consequential than others.

The more we learn about genomes, the more we notice the variability and complexities embodied in each one. Not only do some evolutionary lineages change faster than others; even within a single lineage, different genes and different regions of single genes can evolve at different rates. These variable rates of change support the view that genomes are a mix of functionally consequential and functionally neutral mutations.

Selection, then, does not always select the very fittest, because chance can also influence which genotypes survive and reproduce. Biologists sometimes distinguish between natural selection and “sorting,” which refers not to selection correlated with adaptation, but to selection of a quirkier, chancier sort. Consider, for example, an oyster exuberantly spawning a million eggs in a reproductive season, with only one or two surviving to adulthood. Many of the “loser” eggs undoubtedly fail to survive because of defects that decrease their chances for larval survival in open water or their chances of later completing the physiologically stressful process of metamorphosing into young oysters. But just as surely, among the perished embryos and larvae, there were plenty of physiologically healthy individuals—conceivably even a few larvae more healthy than the survivors—that died not because of defects, but because they had the bad luck to end up too close to the mouth of a predator. In this case, selection involves not just the demise of the less fit but also the demise of the unlucky.

Such imperfect coupling of selection and adaptation reflects the number and complexity of factors affecting the lives of organisms; our understanding of those factors is imperfect as well. Only a naïve biologist would expect 100 percent correspondence between selection and what is, or more accurately what appears to be, adaptation.

But could random factors play an even greater evolutionary role in some cases? The most plausible examples of selection-imperfectly-coupled-with-adaptation are generally those that, like selection pressures on oyster larvae, might be considered among the most competitive: bottleneck situations where the number of surviving individuals is winnowed by 99 percent or more. A hurricane decimates a fish population in a coral reef; of millions of tapeworm eggs cast out of a dog’s intestine, only one or two might encounter a new animal host in which they can survive. Survivorship in such situations is determined largely by chance. Consider a bottleneck situation on greater time, geographic, and taxonomic scales: events of mass extinction. In these events, not only do large numbers of individuals of single species perish, but all individuals of many species perish; whole lineages— genera, families, orders—can cease to exist. At the end of the Permian period (the end of the Paleozoic era) some 250 million years ago, 75 to 96 percent of all animal species disappeared. Entire groups of reef-dwelling invertebrate animals disappeared. Even reefs themselves were absent for the next 7 to 8 million years. The end of the Cretaceous period (the last period of the Mesozoic era), 65 million years ago, saw the extinction not only of dinosaurs, but also of several species of marine reptiles, and terrestrial plants and mammals, all ammonites (shelled relatives of squid), and many other marine invertebrates, protozoa, and algae.

Adaptive deficiencies cannot explain all these extinctions. Paleontologists Anthony Hallam and Paul Wignall noted that the second half of the Permian extinction “evidently came as a surprise to most of the world’s biota,” with protozoans, sponges, and land plants still diversifying almost to the time of their extinction. However much dinosaur diversity or abundance may have decreased before the end of the Cretaceous, there is no hint that dinosaurs perished from gross maladaptation; as the paleontologist David Raup once wrote, “No Mesozoic biologist could have predicted [the dinosaurs’] demise.”

The likelihood that asteroid impacts set off the Cretaceous extinction (and perhaps others) bolsters the view that mass extinctions are of a scale so colossal, and of causes so drastic, that they largely transcend questions of species fitness. Adaptation is surely part of the story, especially for the immediate survivors whose descendants we live with today; but for the species that became extinct, the story was the demise of the ecologically or geographically unlucky, not only selection against the unfit.

Conway Morris sees “the emergence of particular evolutionary properties,” such as intelligence, as the most important aspect of evolution; the genetic lineage in which new properties emerge is, to him, less important than the property itself. I disagree. The character of specific lineages does matter. The genetic context in which intelligence evolves, and the context in which intelligence operates can, in fact, matter a lot. Couldn’t the impact of intelligence differ in organisms with life styles that differ from ours? Might not such organisms affect the biosphere in different ways than we do? In mass extinctions, the body plans and genomes of many major lineages were lost forever. Each of these perished lineages offered a unique genetic palette with which selection could work. Stephen Jay Gould was right to suggest that the survival of such now-extinct lineages could well have generated a different evolutionary outcome than that which we see today.

Conway Morris and others argued against Gould that, no matter which genetic lineages survived mass extinctions, the mechanical and functional constraints on an organism’s design would still lead, inexorably, to a flora and fauna functionally (if not genetically) similar to those we see today. “The routes are many, but the destinations few,” writes Conway Morris in Life’s Solution.

This is, to me, a bleak view of the evolutionary process; it is also, at best, incomplete. The answer to the question “Life: Cosmic Accident or Cosmic Destiny?” is that evolution is not merely accident or destiny, but both. It is also a story of both convergent and divergent destinations.

Evolutionary history is reflected not only in the shared designs and genetic heritage among species, but also in the uniqueness of each species. The richly diverse plant life of the Arizona desert does show intriguing convergence in the architecture of its cacti, euphorbs, and agaves; but to describe the evolution of these plants in terms of only one or several “destinations” misses much of their message. Even these convergent plant species still differ among themselves, not just in trivial details, but in ways that really matter: in the length of their roots; in their sizes, shapes, growth rates, and life spans; in the shapes and blooming periods of their flowers; in the organisms that pollinate them, eat them, or parasitize them. And hundreds of other plant species inhabit the Sonoran Desert: ocotillos, for instance, thrive alongside the iconic saguaro cacti, offering distinct, but equally successful, solutions to desert life. The diversity of Arizona desert flora tells us not that there is one best way for a plant to live in a desert, but that there are many ways to live there. The several million species of organisms on this planet offer us a similar message: there is on earth not just one evolutionary destiny or one solution, but millions.

However stridently some may insist that religion is at odds with evolution, it is a mistake to characterize the general debate between purpose and chance only in religious terms. Among scientists, no absolute correlation exists between religious viewpoint and evolutionary philosophy: atheists, agnostics, and believers populate both sides of the debate. While Conway Morris apparently sees theological overtones in the ways he finds the universe “strangely fit to purpose,” his allies are by no means all religiously inclined. Siding with Conway Morris on the contingency issue is Richard Dawkins, a proud, indeed proselytizing, atheist. Though Dawkins surely disavows the mystical aspects of Conway Morris’s evolutionary vision, he shares with him an abiding faith in adaptation as the prevailing evolutionary force; and he has argued fiercely alongside him against historical contingency as a major influence on evolutionary history.

On the other side is, or was, Stephen Jay Gould, who argued for contingency as a significant player in the history of life. An agnostic, Gould nevertheless knew an impressive amount about Judeo-Christian religious tradition, and was respectful of it. Another scientist acknowledging evolutionary happenstance is George Coyne, an astrophysicist, Jesuit priest, and director of the Vatican Observatory. A man of deep and explicit religious faith, Coyne nevertheless has shown impatience with human-centric, universe-made-for-man perspectives, telling Wired magazine in 2002 that “to imagine a Creator twiddling with the constants of nature is a bit like thinking of God as making a big pot of soup,” a view he characterized as a regression to archaic notions of a “watchmaker God.” For Father Coyne, no conflict exists between God and a world of evolutionary surprise.

Thus, debates about design vs. chance in nature are not, finally, about religion vs. science, scientists vs. nonscientists, or evolution with or without a Creator. They are instead about something deeper: a fear among many of us that human life has no meaning if chance has played a role in our origins, no value if the destiny of evolution is not solely about us. But neither happenstance in our history nor evolutionary kinship with fellow creatures need rob our lives of meaning. These factors enrich our lives by offering us new dimensions of understanding; and by acknowledging our evolutionary relatives, we should also feel less alone.

Perhaps several million years from now (or perhaps even now, on another planet in another galaxy, or, if the string theorists are right, another universe), a wise, soft-bodied, eight-limbed paleopsychologist, unburdened by pesky vertebrae and clogged sinuses (but surely with other limitations of its own), will take as its life’s work the contemplation of the psyche of that once-dominant, impressively clever but comically arrogant, touchingly insecure, sadly self-destructive species, Homo sapiens. Perhaps that scholar and its fellow octopods will have the confidence about their place in the universe, and the appreciation of their many interdependencies with their species-neighbors, to enable them to use their intelligence to foster the health of the biosphere that will support them all. Now that would be evolutionary progress.

Mary Beth Saffo is a biologist at work on a book about parasitic and beneficial symbiosis, tentatively titled "Lives of the Infectious and the Infected."

Comments are closed for this post.