Adaptationist Claims – Conceptual Problems

Introductory article Article Contents . Introduction . Telling Adaptationist Stories . Atomizing Organisms

Christopher D Horvath, Illinois State University, Normal, Illinois, USA

. Conflating Adaptive with Adaptation . Alternative Evolutionary Explanations

Adaptationism is the name for a general approach in evolutionary biology that emphasizes the importance of natural selection above all other evolutionary processes. In 1978, Gould and Lewontin published a now classic critique of adaptationism, launching a fierce debate about both the empirical and conceptual adequacy of this approach.

Introduction Biologists generally think of individual organisms as collections of characteristics called traits. For example, the platypus has a distinctive set of traits that include egglaying, poisonous spurs on its rear legs, and an unusual duck-like bill that carries organs sensitive to small electric fields. However, any given organism will possess a huge number of traits. One pressing problem for biologists is to determine which of an organism’s traits are important and worthy of study. One obvious way to decide is to determine which traits are important in the organism’s evolution, ecology and development. These are the traits that will be worthy of study. One clearly important set of traits is the set of traits that increase the relative fitness of the organisms that bear them and are thus favoured by natural selection. In other words, these traits give their bearers some advantage in the competition for survival and reproduction. Biologists call such traits ‘adaptive’. A trait that exists as a consequence of being or having once been adaptive is called an ‘adaptation’. For example, the eye-blink reflex in humans is adaptive; it serves to protect the eye from injury by foreign objects. The eye-blink reflex exists in humans today because it protected the eyes of our ancestors and so increased their fitness. Thus, the eye-blink reflex is an adaptation for protecting the eye. (see Fitness.) (see Fitness: philosophical problems.) (see Adaptation: study.) (see Adaptations: meanings.) Adaptationism is the name for a general approach in evolutionary biology that emphasizes the importance of natural selection above all other evolutionary processes. The central tenant of adaptationism is that every adaptive trait an organism possesses must be an adaptation. In other words, any trait an organism has that is currently useful must be there because natural selection favoured it. According to adaptationism, natural selection is the most important evolutionary process and it alone can explain the existence of the complex adaptive features found in

. Testing Adaptationist Hypotheses

most organisms. (see Adaptation and natural selection: overview.) (see Natural selection: introduction.) In their now classic critique of adaptationism, Stephen Jay Gould and Richard Lewontin compared adaptationists to Voltaire’s Dr Pangloss, who believed that everything in the world was designed by a loving God to be the best it could possibly be. Gould and Lewontin accused adaptationists of having the equally unrealistic view that every trait an organism possessed was there because natural selection designed it to be the best possible solution to some problem presented to the organism by its environment.

Telling Adaptationist Stories Perhaps the best way to see what making adaptationist claims entails is to examine how an adherent of the approach would solve an evolutionary puzzle about the features of some organism. Adaptationism is a powerful force in contemporary evolutionary psychology, so let us pick an example from this field. Margie Profet’s explanation of pregnancy sickness (morning sickness) in human females provides a classic example of adaptationist thinking. Why is it that many women experience ‘morning sickness’ during the first few weeks of their pregnancies? The first step in the adaptationist strategy is to show how pregnancy sickness is adaptive. Profet’s theory is that pregnant women become hypersensitive to certain odours and that this hypersensitivity and accompanying nausea helps them to avoid eating foods that are normally harmless to adults, but which may harm the developing fetus even in small quantities. She then goes on to provide an ‘adaptive story’ to explain why pregnancy sickness is common among modern human females: our ancestors who possessed the genes for this hypersensitivity ingested fewer harmful foods and thus had more healthy offspring. This trait then spread through the human population by natural selection. Thus, pregnancy sickness is an adaptation for avoiding the ingestion of teratogens and mutogens during early pregnancy.

ENCYCLOPEDIA OF LIFE SCIENCES © 2001, John Wiley & Sons, Ltd. www.els.net

1

Adaptationist Claims – Conceptual Problems

Atomizing Organisms As Gould and Lewontin argued in their critique, adaptationists make four principal kinds of mistakes. First, they seem to see organisms as collections of separate parts, each of these parts with its own independent evolutionary explanation. They believe that an organism can be divided into discrete traits where each of these traits is optimized to the organism’s environment. While it may be useful to describe organisms in terms of their characteristic traits or to use these traits as a way of placing organisms into taxonomic groups, in reality organisms are not collections of discrete traits, each optimally designed for its particular function. (see Philosophy of biological classification.) (see Function and teleology.) Part of the issue here is how to divide up organisms into parts. Male mandrills (a large Old World monkey) have electric blue muzzles and matching blue fur on their genitals and buttocks. Are these blue patches of fur separate traits with separate evolutionary histories, or is the monkey’s coloration a single trait? At first sight, the human chin appears to be a trait. In fact, some people have argued that a ‘strong’ chin is adaptive in human males as a marker of genetic quality. But, as Gould and Lewontin point out, the human chin cannot be an adaptation. It is not a phenotypic feature that has its own evolutionary history. The chin is formed as a result of changes occurring in the two bony structures that make up the lower jaw. Changes to these structures may be due to natural selection, but the chin is simply a by-product of these changes. Even if one could find a principled way to divide the organism into its constitutive traits, the adaptationist treats each trait as if it had an independent evolutionary history. Clearly, there are some traits that may evolve relatively independently of the rest of the organism. The beaks of Darwin’s famous finches seem to evolve in response to selection without other major changes in the organism’s phenotype. Nevertheless, as Gould and Lewontin argue, it is doubtful that all (or even most) traits behave this way. Organisms are not collections of independently optimized traits. Instead, an organism’s overall phenotype represents a compromise between competing demands on all the different aspects of that phenotype. (see Darwin’s finches.)

Conflating Adaptive with Adaptation The second major conceptual problem with adaptationist claims is that they conflate being adaptive with being an adaptation. Adaptationism assumes that any useful trait exists because it has been selected for. The actual relationship between adaptive traits and adaptations is much more complex. Some traits may exist because they were selected 2

for at some point in the evolutionary history of the organisms that bear them, even though those traits are no longer adaptive for those organisms. For example, having hair on their bodies may have helped keep our ancestors warm and hence may have contributed to their reproductive success. If so, then having hair on our bodies is an adaptation; however, given the invention of clothing (not to mention space heaters and thermal insulation), body hair is no longer adaptive. (see Thermoregulation in vertebrates: acclimation, acclimatization and adaptation.) More importantly in this context, the converse situation also exists. There are traits that are currently adaptive that were not selected for and exist merely as a side effect of the evolution of other parts of the organism. The human chin (discussed above) is one example of this phenomenon. Another, more exotic, example can be found in the spotted hyena. The female spotted hyena has a hypertrophied clitoris that closely resembles the penis of male hyenas. The females use this organ in complex greeting ceremonies. The spotted hyena’s hypertrophied clitoris is thus adaptive. However, the hyena’s clitoris did not evolve for this or any other function. It is merely the side effect of selection for aggressiveness in female hyenas and the increased exposure to androgenizing hormones that accompanies increased aggression. According to adaptationism, all of the adaptive features of an organism must be adaptations, but adaptations for what? The current adaptive effects of a trait may not be the same effect for which it was originally selectively favoured – the effect that made it an adaptation. One of the most often cited examples of such a trait are feathers. Current theory holds that the ancestors of modern birds that first developed ‘feather-like’ scales were more fit than those with more typical scales because the feather-like scales were better for thermoregulation. Many generations later, after a great many other evolutionary changes had occurred in this lineage, the descendants of these early birds found their feathers to be useful in flight. Hence, feathers were originally an adaptation for temperature regulation, and only much later in the history of the lineage became adaptations for flight. When the adaptationist looks at the current utility of feather, she might give an adaptationist story that emphasized the adaptive role of feathers in flight, ‘modern birds have feathers because birds with feathers had an advantage in flight over birds without feathers’. Such an explanation would not be false, but it would be incomplete. The complete story would have to involve an account of the changing adaptive role played by feathers throughout the evolutionary history of bird lineages. (see Dinosaurs and the origin of birds.)

Alternative Evolutionary Explanations The third major error made by adaptationists is to ignore explanations other than natural selection for the main-

Adaptationist Claims – Conceptual Problems

tenance of certain traits within lineages. An adaptationist might look at the human forearm and be tempted to offer an adaptive explanation for why it contains two bones instead of one. A much more plausible explanation would rely on the concept of phylogenetic constraint. Humans belong to a large group of vertebrates called the tetrapoda; all have a typical four-limb body plan. The two-boned forelimb design is a highly conserved characteristic of this lineage. That is to say, members of the tetrapod lineage retain the two-bone design by default. Future members of the lineage will also have the two-bone design unless there is strong selection driving forelimb anatomy in some other direction. The evolutionary process that results in the existence of two bones in the human forearm is the conservation of a trait within our phylogenetic lineage and not natural selection for that design. (see Tetrapod walking and running.) The bottom line here is that natural selection is not the only evolutionary process going; nor is natural selection all-powerful. Many biological and physical processes limit the results that can be produced by natural selection. In addition to the phylogenetic constraints discussed just above, there are also allometric constraints that limit the size and shape of organisms. For some trees, the optimal solution to being shaded out by your neighbours might be to grow taller. However, there are physical limits placed on the height to which trees can grow given that they must be made of plant cells with relatively rigid cell walls. The optimal solution may be physically impossible. The genetic architecture of organisms often constrains genetic recombination thus limiting the amount of genetic variability on which natural selection can operate. There are also constraints placed on natural selection by the highly conservative and integrated nature of development. (see Evolutionary ideas: the modern synthesis.) (see Adaptation: genetics.) It is impossible to tell a priori how important any of these constraints might be in any given lineage. Those who criticise adaptationism are not suggesting that these alternative evolutionary processes explain all, or even most, of the adaptive traits found in the world. Nevertheless, these processes do exist and their relative importance in explaining any given trait is an open question requiring some empirical investigation. (see Adaptation and constraint: overview.)

Testing Adaptationist Hypotheses Finally, a fourth major problem with the adaptationist approach is that adaptationist explanations are generally unfalsifiable. Falsifiability is the hallmark of good scientific hypotheses and separates sciences like biology and physics from pseudo-sciences like creationism and astrology. A

good scientific hypothesis will offer some way of using empirical evidence to determine its validity. In their critique, Gould and Lewontin accuse adaptationists of telling ‘just so stories’ about why a trait was selectively favoured at some point in an organism’s distant past. The adaptationist looks at a given trait and ‘makes up’ a plausible story about how that trait might have evolved in that lineage and then accepts this story as an explanation. The best adaptationist explanations take into account what we know about the actual ecology and environment faced by the organisms in questions during the relevant time period. The problem is that we most often know very little about the physical environment and even less about the ecological environment faced by ancestral organisms. Adaptationists are free to make all sorts of claims about these ancestral environments because we have very few ways of checking to see if what they are claiming is true. (see Palaeoecology.) It is possible, at least in principle, to test some individual adaptationist hypotheses. Take the example of pregnancy sickness discussed above: one could easily check to see if the foods that seem to cause an adverse reaction in pregnant women actually contain teratogens. There may be enough evidence about our ancestors to assess whether or not these same food items were a substantial part of their diets. One could also check to see whether there is any reasonable hypothesis that explains how this phenomenon operates at a physiological level within individual human females. Answers to these questions would provide evidence for or against Profet’s adaptationist hypothesis. However, testing individual adaptationist hypotheses does not provide a test for adaptationism itself. If the tests were to suggest that a given adaptationist explanation was false, then the standard practice would be to offer another adaptationist explanation in its place. Adaptationism is unfalsifiable, because adaptationists never actively consider nonadaptive alternatives.

Further Reading Amundson R (1994) Two conceptions of constraint: adaptationism and the challenge from developmental biology. Philosophy of Science 61: 556–578. Brandon R (1990) Adaptation and Environment. Cambridge: MIT Press. Ellstrand N (1983) Why are juveniles smaller than their parents? Evolution 37: 1091–1094. Gould SJ and Lewontin RC (1978) The spandrels of San Marcos the Panglossian paradigm: a critique of the adaptationist programme. Proceedings of the Royal Society of London B 205: 581–598. Orzack S and Sober E (2001) Optimality and Adaptationism. Cambridge: Cambridge University Press. Profet M (1992) Pregnancy sickness as adaptation: a deterrent to maternal ingestion of teratogens. In: Barkow J, Cosmides L and Tooby J (eds) The Adapted Mind: Evolutionary Psychology and the Generation of Culture. New York: Oxford University Press.

3