While reading Denis Noble’s contribution to this Dialogue and reviewing my own, I found myself wondering what a reader encountering the ideas for the first time might be thinking. I worry that such a reader might feel bewildered and ultimately bored by who said what and whether it was well received or ignored. Can’t we just move ahead with the best of our current knowledge?
More interesting is the question of why science moves rapidly along some fronts and slowly along others. Everyone can agree that the pace of scientific change is highly variable. If we can diagnose the causes of variability in the past, then perhaps we can accelerate the pace of scientific change in the future. It is in this spirit that the history of evolutionary thought becomes interesting and relevant to those encountering the subject for the first time.
It is important to keep in mind that Professor Noble and I are defining neo-Darwinism in different ways. My definition is based on a four-fold distinction made by Niko Tinbergen in 1963, which states that a fully rounded evolutionary perspective requires paying attention to function, mechanism, development, and phylogeny. “Tinbergen’s four questions” can be mapped onto a two-fold distinction between ultimate and proximate causation made independently by Ernst Mayr in 1961.
Roughly, Tinbergen’s “function” and “phylogeny” questions map on to Mayr’s “ultimate causation,” and Tinbergen’s “mechanism” and “development” questions map onto Mayr’s “proximate causation.” The claim I defend is that Tinbergen’s four questions remain a concise description of a fully rounded evolutionary perspective and that to suggest otherwise is a distraction.Noble defines neo-Darwinism in terms of a number of positions that emerged during the twentieth century: that genes are the only mechanism of inheritance; that genetic variation is random with respect to what is selected; that germ cells are isolated from somatic cells; and so on. Positions such as these have a way of hardening into dogmas that resist change for decades, and to some extent still today. I do not contest these facts and I am a veteran of another position that hardened into a dogma: that natural selection always operates at the level of the individual and/or the gene.
It is fine for us to define neo-Darwinism in different ways. As long as we are clear about our differences, then we are like two mountaineers viewing the same mountain from different angles to plan an ascent. I like my angle because it has great current utility. If Tinbergen’s four questions remain good enough, then they can be easily taught and kept in mind by beginner and expert alike. Against this background, I will review Professor Noble’s 10 reasons why from his perspective “neo-Darwinism is not enough,” with three questions in mind. First, are his 10 reasons legitimate? Second, do they challenge my claim that Tinbergen’s questions are good enough? Third, why didn’t scientific inquiry proceed more quickly and smoothly for these topics?
1. Major diseases still plague humanity. Professor Noble correctly notes that the Human Genome Project largely failed to deliver on its grandiose claims. It turns out that knowing the DNA sequence of organisms does not provide sufficient information to explain more than a small fraction of variation for phenotypic traits such as body height or susceptibility to a disease. This is because most phenotypic traits are highly polygenic and influenced by gene-gene and gene-environment interactions operating though complex developmental and physiological systems.
Does this pose a challenge to Tinbergen’s four questions? Not in the least, since they call upon us to study mechanisms and development. Then why were the architects of the Human Genome Project so naïve as to think that phenotypic traits map directly onto genes? In part because they were selling something to Congress. In part because of the appeal of reductionism, which manifests itself for many topics in science, not just evolution. I have noticed for the study of ecology, for example, that holistic descriptions of the tapestry of nature in the 1950s gave way to simple models of single species dynamics and multispecies interactions. These models were somewhat explanatory, but also revealed their own weaknesses. Complications were added and now nature is described as a grand tapestry, after all. Of course, the current description is mechanistically far richer than the earlier description.
Is it necessary for the study of proximate mechanisms to pass through a simplistic phase before embracing more complex explanations that were always known to have existed? I don’t know, but I agree with Professor Noble that the study of physiological systems was neglected and deserves the kind of support that the Human Genome Project received.
2. Privileging any one level in biological systems cannot be justified. I worry that Professor Noble is unduly influenced by a few provocateurs such as James Watson and Richard Dawkins. Watson’s statement “there are only molecules—everything else is sociology” is tantamount to saying, first, that the mechanism question is the only one worth asking, and, second, that the mechanism question can be answered entirely at the molecular level. I agree with Professor Noble that both of these claims are stupid, but I also don’t think that Watson’s quip represents neo-Darwinism or probably even Watson in more sober moods.
There are two major forms of holism that are already part of the neo-Darwinian synthesis, even as formulated by Professor Noble. The first is ultimate causation (=Tinbergen’s function and history questions). The holistic statement “the parts permit but do not cause the properties of the whole” is literally correct when environmental forces operate on heritable variation. That is why the distinction between proximate and ultimate causation is so insightful.
The second form of holism is purely mechanistic and establishes that higher-level physical entities, such as molecules, have properties that cannot be reduced to the properties of their parts, such as their constituent atoms. This point can be made for compounds such as salt and water without requiring the biological examples that Professor Noble provides. It is so firmly established that one article (Sober 1999) begins, “If there is now a received view among philosophers of mind and philosophers of biology about reductionism, it is that reductionism is mistaken.”An important lesson to learn is that a scientist such as James Watson can be brilliant and well-read in some respects and painfully ignorant in other respects. A distinguished scientific reputation is not (or should not be) a license to hold forth on all subjects.
To conclude, Professor Noble’s second critique is valid, but doesn’t fairly represent the neo-Darwinian synthesis, even as he defines it, and affirms rather than challenges the four-question perspective.
3. The gene-centric view has damaging consequences. Alas, I’m afraid I must agree with Professor Noble that he is not a social historian and is making statements on a par with Watson’s quip about molecules. To begin, some of the disturbing implications of evolution began with Darwin, not neo-Darwinism. The Christian worldview imagines the universe as harmonious from top to bottom, from the smallest insect to the stars in heaven. Darwin’s theory explained functional design at the level of individual organisms, but required special conditions (group selection) to explain functional design even at the level of single-species social groups, much less higher levels of nature such as ecosystems and the biosphere. Gene-centrism had nothing to do with these and other disturbing implications of Darwin’s theory for the Christian worldview.
While Darwinism was genuinely threatening to some worldviews, the idea that it unleashed a plague of toxic policies justifying the strong taking from the weak is largely false. I speak with authority on the subject, having organized a special edition of “This View of Life” titled “Truth and Reconciliation for Social Darwinism.” The conjecture that evolutionary theory’s turn toward gene-centrism had additional pernicious effects is even more far-fetched. I encourage Professor Noble to read the special edition before writing anything more on this subject!
The idea that we are born selfish and must be taught altruism goes back millennia and has been expressed in many different forms, including the Christian concept of original sin. It’s true that Dawkins and others such as G.C. Williams, Richard Alexander, and Michael Ghiselin gave the idea an evolutionary spin in the 1960s and 1970s. It’s probably not a coincidence that this development in evolutionary biology coincided with the rise of methodological individualism in the social sciences, rational choice theory in economics, and the rise of individualism in popular culture. The trend in evolutionary biology was more likely a manifestation of these other trends than the cause. Nor was gene-centrism required to create an individualistic worldview. Friedrich Hayek, one of the main architects of neoliberal economic thought, was an early adopter of cultural group selection theory, which did not prevent him from becoming individualistic in his own way.
These topics have little do to with neo-Darwinism, even as defined by Professor Noble, and they don’t threaten the four-question perspective, but they do shed some light on why scientific controversies can be protracted—because they are thoroughly entwined with non-scientific, cultural worldviews. Please see my own essay for more on multilevel selection as a protracted controversy.
4. The gene-centric view resists new findings. I am puzzled by the first paragraph of this section. Far from resisting a new finding, a supporter of neo-Darwinism accepted the inheritance of acquired characteristics, but for Professor Noble this was a contradiction in terms.
The reason that directed forms of evolution can be reconciled with neo-Darwinism is because they can be shown to evolve from undirected forms of evolution. Once this point is understood and mechanisms are identified, then there is no barrier to accepting directed forms of evolution from a neo-Darwinian perspective. Individual learning and cultural evolution provide clear examples of directed evolutionary processes. If genetic and epigenetic processes are also discovered, then they will be of the same type.
That said, I agree with Professor Noble that many evolutionary biologists are dogmatically opposed to anything that smells like Lamarckism, just as others are dogmatically opposed to anything that smells like group selection. Why should this be? In part because they are often taught about these subjects in a dogmatic fashion. Much as we might want science education to be open-minded, it often takes the form of identifying what is taboo and what is permitted, even at the college and post-graduate levels. This is a problem for science education in general, which is by no means restricted to the gene-centric view.
5. The gene-centric review claims parsimony. I am familiar with this claim because parsimony played a large role in the rejection of group selection. The idea was that individual-level selection is somehow simpler than group-level selection and therefore preferable in the absence of other deciding information. I agree that this is an extremely weak argument for most topics in science, as Elliott Sober has recently recounted in detail. It is especially weak within evolutionary theory, where adaptations are expected to be more like what a tinkerer, rather than an engineer, would build.
While I side with Professor Noble on the topic of parsimony, I still must object to this passage: “Whole areas of economics, sociology, and philosophy are based on interpretations of selfish gene viewpoints.” I think that this passage greatly overstates the influence of gene-centric thinking in these other disciplines. See my interview with the sociologist Russell Schutt titled “Why Did Sociology Declare Independence from Biology (and Can They Be Reunited)?” (which is part of the Truth and Reconciliation Special Edition), for a more informed view of that discipline in relation to evolutionary theory.
6. The gene-centric view claims to settle the question of Lamarckism. In this and other sections, Professor Noble reviews the work of figures such as James Mark Baldwin, C.H. Waddington, Barbara McClintock, and Lynn Margulis, to which we could add figures such as Sewall Wright and concepts such as niche construction and developmental systems theory. All of these people and concepts were influential enough to be remembered—what student of evolution hasn’t seen an image of Waddington’s developmental landscape or Wright’s adaptive landscape?—but somehow failed to occupy center stage.
Why should this be?
In part, because they embraced the notion of complex interactions, which are inherently more difficult to study than more simple interactions. I have some sympathy with parsimony as a research strategy, which attempts to explain as much as possible with simple models before resorting to more complex models. It is instructive to compare the three main pioneers of theoretical population genetics: Ronald Fisher; J.B.S. Haldane; and Sewall Wright. All three were gene-centric, the difference being that Wright was more drawn to study complex genetic interactions that result in multiple local equilibria (his shifting balance theory), whereas Fisher and Haldane began by modeling genes with additive effects. Fisher and Haldane were arguably able to make more rapid progress by picking the “low-hanging fruit” with their simple models. Hence they had a greater influence at first, but once the low-hanging fruit had been picked, then Wright’s work became increasingly important. If all of this takes place in the span of a few decades, then it is hard to fault the scientific process for being too biased or too slow.
Another point is that if it’s possible to make an end run around complexity, then by all means one should do so. Adaptationist thinking (Tinbergen’s function question) is one example of an end run. By assuming heritable phenotypic variation, it makes progress without needing to understand the underlying mechanisms. As another example, imagine comparing the phenotypes of two genotypes that differ by only a single mutation. Each allele influences the phenotype through a complex web of developmental and physiological interactions, but these interactions can be ignored if we merely want to select the mutant phenotype. If rapid progress along these lines took place in the mid-twentieth century, with topics such as evo-devo only coming into their own during late in the century, should we be too surprised?
Another point is that a few decades is not a long time. Science is a process of constructive disagreement and several rounds of hypothesis testing might be required to dislodge someone’s firmly held view. Professor Noble states that “the 1956 version of the Central Dogma had to be substantially revised in the 1970’s when it was discovered that the step from DNA to RNA is in fact reversible.” Two decades is an extraordinarily short time for this kind of scientific progress!
7. The gene-centric view claims that epigenetic inheritance is short-lived. Scientists can be conservative or progressive, much like politicians or religious believers. Some will rush to say “This changes everything!,” while others will never change their mind, especially if they have received a dogmatic education. This is true for all topics in science and is not restricted to evolutionary theory.
If we think of epigenetic inheritance as a form of multigenerational phenotypic plasticity, then it is helpful to review what we know about intragenerational phenotypic plasticity.
The function of phenotypic plasticity is to track environmental changes during the lifetime of the organism. The exact form of phenotypic plasticity, therefore, depends upon the pattern of environmental changes. Arctic mammals change their coat color with the season. A turtle withdraws into its shell when a predator approaches and comes back out when the predator leaves. Some forms of phenotypic plasticity are closed (involving a fixed repertoire of behaviors), while others are open (involving a variation and selection process such as operant conditioning).
For the purpose of argument, let’s assume that there are no mechanistic barriers preventing the phenotypic state of the parent to be transmitted to its offspring. If so, then we’d expect the same diversity of patterns for transgenerational phenotypic plasticity that we see for intragenerational phenotypic plasticity. The duration of phenotypic change would be calibrated to the frequency of environmental change for each particular trait.
According to Jablonka and Lamb, many forms of transgenerational phenotypic plasticity mediated by epigenetic effects are relatively closed, such as the fast and slow life history strategies that are expressed in rats depending upon how much they are licked by their mothers. These hold little potential for cumulative evolutionary change. However, behaviors that are learned by operant conditioning during one generation and transmitted across generations by various forms of social learning are more open-ended and cumulative. Forms of symbolic thought that are distinctively human have evolved into a full-blown inheritance system that operates in parallel to the genetic inheritance system.
As I have emphasized elsewhere in this Dialogue, some of the most radical claims that are being advanced for genetic and epigenetic mechanisms are similar in kind to claims that have already been established for individual learning and transgenerational cultural evolution. Going back to basics and centering evolution on the concept of heredity, not merely genes, is revolutionary against the background of gene-centric evolutionary theory, a point on which Professor Noble and I agree. Tinbergen’s four questions are needed for everything that counts as an evolutionary process, no matter what the particular mechanism of inheritance.
8. The gene-centric view claims genetic change is always random with respect to function. Much of what Professor Noble writes in this section affirms this view, adding that randomness is also controlled and incorporated into function. Behavioral examples can be added to the physiological examples. Almost all animal and human decision-making involves a component that is random with respect to what is selected. The same is true for evolutionary algorithms in computer science. Where else can novelty come from? None of this is a radical departure from neo-Darwinism, even as defined by Professor Noble, nor does it challenge the four questions perspective.
9. The gene-centric view claims neo-Darwinism is obviously and necessarily true. Neo-Darwinism, as defined by Professor Noble, is a collection of empirical claims. Each claim is acknowledged by all to be empirically falsifiable, the quotes by Weissman and Dawkins notwithstanding. When Weissman chopped off the tails of mice, he was prepared to accept the results if the next generation of mice had shorter tails.
A neo-Darwinist can accept Lamarckian inheritance while remaining a neo-Darwinist if the mechanism of Lamarckian inheritance can be shown to evolve by a neo-Darwinian process. Since that is the case for all known Lamarckian inheritance mechanisms, then Lamarckian inheritance has been reconciled with neo-Darwinism. Some neo-Darwinists can be faulted for failing to acknowledge Lamarckian inheritance or for trying to marginalize its importance, but it is not helpful for Professor Noble to declare that Lamarckian inheritance lies outside the orbit of neo-Darwinism by definition.
10. The gene-centric view appeals to authority. I began my own essay by saying that I was an odd choice to defend the position “Is neo-Darwinism enough?”, since I am a well-known opponent of selfish gene orthodoxy and a booster of the ideas that sail under the name “Extended Evolutionary Synthesis.” Hence, I admire Professor Noble’s systems approach to physiology and share his holistic worldview.
Nevertheless, when I read his essay in detail in preparation for writing this response, I became increasingly unhappy with his critique of neo-Darwinism as he chooses to define it. His parting shot is especially cheap. Professor Noble seems to have worked himself up into such a “us vs. them” lather that he can’t even grant that his opponents also function as scientists and, as a group, don’t appeal to authority any more or less than other scientists.
In my own essay, I chose to defend the position that Tinbergen’s four questions provide a concise description of a fully rounded evolutionary perspective that remains good enough for student and expert alike. To claim otherwise is a distraction, especially since broad swaths of scientific inquiry do not employ a fully rounded approach, even in the biological sciences.
In this response to Professor Noble’s essay, I attempted to answer three questions. First, how successful is his critique of neo-Darwinism, as he chooses to define it? Second, does his critique challenge neo-Darwinism as I define it (Tinbergen’s four questions)? Third, what can we learn from the history of neo-Darwinism about the factors that accelerate or impeded scientific progress?
My conclusion for the first question is mixed. I am broadly sympathetic with Professor Noble’s systemic and holistic view of evolution, but many aspects of his critique of neo-Darwinism, as he defines it, are highly problematic.
My conclusion for the second question is that Tinbergen’s four questions emerge unscathed. They are not challenged, and are frequently affirmed, by Professor Noble’s 10 points.
My conclusion for the third question is that the history of neo-Darwinism, as defined by Professor Noble, includes many examples of science progressing more slowly than we might like. Here is a list:
- Selling science to funders requires false and boastful claims
- The appeal of reductionism
- Scientific worldviews are bound up with cultural worldviews
- Science education that is more dogmatic than we might like
- Appreciation of complex interactions comes after discovering the limitations of explanations based on simple interactions
- A few decades is not a long time for science as a process of constructive disagreement
What’s striking about the list is that none of the items has much to do with neo-Darwinism per se. All are likely to impede progress for any scientific subject.
Although I have been critical for Professor Noble in some respects, I agree with him that evolutionary science will make major advances during the twenty-first century, especially for the study of evolution in relation to human affairs. As I see it, this is because the fully rounded, four-question approach will become much more widespread than it is now.
1. N. Tinbergen, “On aims and methods of ethology,” Zeitschrift für Tierpsychologie, 1963, 20: 410–433.
2. E. Mayr, “Cause and Effect in Biology,” Science, 1961, 134(3489): 1501–1506.
3. D.S. Wilson, “Holism and Reductionism in Evolutionary Biology,”Oikos, 1988, 53: 269–273.
4. E. Sober, “The Multiple Realizability Argument Against Reductionism,” Philosophy of Science, 1999, 66: 542–564; p. 542.
5. G.C. Williams, Adaptation and Natural Selection: A Critique of Some Current Evolutionary Thought. Princeton: Princeton University Press, 1966.
9. G.C. Williams, op. cit.; for a critique, see E. Sober and D.S. Wilson, Unto Others: The Evolution and Psychology of Unselfish Behavior. Cambridge, MA: Harvard University Press, 1998.
11. E. Jablonka and M. Lamb, Evolution in Four Dimensions: Genetic, Epigenetic, Behavioral, and Symbolic Variation in the History of LIfe. Cambridge, MA: MIT Press, 2006.
12. N.M. Cameron, D. Shahrokh, A. Del Corpo, S.K. Dhir, M. Szyf, F.A. Champagne, and M.J. Meaney, “Epigenetic programming of phenotypic variations in reproductive strategies in the rat through maternal care,” Journal of Neuroendocrinology, 2008, 20(6): 795–801.
13. R.A. Paul, Mixed Messages: Cultural and Genetic Inheritance in the Constitution of Human Society. Chicago: University of Chicago Press, 2015.