In this post I will summarize my talk and add a few additional comments. Before my talk, I sent out an email that said I was going to arouse people to form a free opinion, rather than try to convince. As will be seen, it was important to explain what counts as “evidence” for a theory and my approach. I begin the talk with a quick summary of the theory’s structure, and some patterns of evolution consistent with it. Below, I will go approximately in order of the slides and review what I said. I do not cover everything I said in the talk, and I also add a few things here.
Most theories are not directly falsifiable. There is a buffer between theories and data provided by models.
If theories are not directly falsifiable, then “withstanding scrutiny” is not evidence for a theory. Then what is evidence for a theory?
This question can be reframed as: what makes us think a theory is correct?
We tend to assume a theory is correct when we learn about it in our textbooks (e.g., Evolution, 4th edition)… when everybody else adopts and uses the same theory.
Thus, the key to understanding evidence for a theory is understanding what makes scientists adopt a theory.
Scientists adopt theories for various reasons: theories (1) help resolve anomalies found under former theories, (2) guide new lines of empirical investigation, (3) allow unification of formerly disconnected bodies of research, and (4) allow scientists to avoid sterile debates or unproductive avenues of research.
Of all the reasons we adopt a theory, the most convincing initially is how it resolves anomalies of a former theory. Focusing on anomalies is useful because workers are typically consumed with the worldview of the current theory. The anomalies are viewable from the former paradigm and provide a bridge between the old theory and the new one.
Although resolving anomalies can provide evidence for a new theory, anyone strongly committed to an old theory can usually make arguments against the anomaly resolutions. They do so either by arguing that the problems were not really anomalies of the former theory, or they question the new solutions. Both are normal and healthy activities that ensure theories are not too quickly modified or surrendered.
In biology, there is a great deal more room for argument on specific points than in physics because of the complexity and ambiguity of the subject matter. Darwin on p. 2 of The Origin stated that he recognized there was not a single point that couldn’t be disputed.
Was Darwin just being modest, or was this true?
In fact, it was true. Among the examples Darwin gave in The Origin, there is not a single “smoking gun” line of evidence for natural selection comparable to say, Eddington’s test of General Relativity. There is some convincing evidence for descent with modification, which is why descent was accepted before natural selection.
Darwin won with a staggering weight of evidence, not a single smoking gun example.
Few of Darwin’s contemporaries were able to impartially weigh the whole body of evidence, or allow it to stimulate them to adopt the theory.
As Darwin predicted, it was younger naturalists who would be capable of impartially evaluating the “evidence” for the theory, i.e., gauging that the theory would be useful as an engine for scientific discovery. One such individual was R. A. Fisher, who recognized that a work of genius like Darwin’s does not easily “lend itself to the scissors.”
At this point, the talk now suggests that the only way to convince an audience would be with weight of evidence. But since it was already necessary to explain what counts as evidence, and how evidence is evaluated in these cases, there is not enough time. It is also necessary to explain another complicating factor, which is the lack of single previous agreed-upon theory of macroevolution.
Darwin had the luxury that during his time, there was a single, widely adopted theory of Independent Creation that served as a backdrop upon which to compare the theory of natural selection. God set Darwin up perfectly. Today, in contrast, there is not a single, widely-accepted theory of macroevolution. Instead, there are many unique and eclectic views on this subject. How then, is it possible to compare a new theory to an old one, as Darwin did?
To answer this question, it is important to categorize existing views on macroevolution. Because the conclusions of the Modern Synthesis still reign supreme in biology, it is useful to start with the views expressed by its authors. These fall into categories of agnostic and extrapolationist.
The agnostic view, adopted by Simpson (1949) and some others, refutes previous macroevolutionary theories of orthogenesis and saltation, but leaves open the possibility of unique macroevolutionary forces and processes.
The extrapolationist view, adopted by Dobzhansky (1951) states that all of macroevolution can be understood without invoking any additionally macroevolutionary forces or processes. This view assumes that macroevolution can be understood by extrapolation of the known microevolutionary forces and processes, including mutation, recombination, natural selection, drift, etc. It assumes natural selection is the only deterministic (non-random, non-teleological) force of evolution.
The extrapolationist view is, however, a bridge to teleological theories of macroevolution, especially as it is applied to adaptive landscapes. Dobzhansky (1951) proposed an adaptive landscape with peaks representing species, where natural selection creates species adapted to their niches. This paved the way for teleological interpretations of natural selection by Williams (1966) and Ayala (1970). It opened the door for theories that invoke selection on long-preserved units, like genes and species, and adherence to “fitness maximization” dogma applied to whole organisms or complex traits.
The authors of the modern synthesis reluctantly embraced the extrapolationist view. I conjecture they embraced it, despite their reservations about overstating their claims, because it excludes the possible “hand for God” in evolution. Creationists constantly search for any chink in the evolutionist’s armor, and use it to justify teaching Creationism in high school science classrooms (e.g., Google “wedge strategy”). The Modern Synthesis authors were cognizant of this and so tried to outwardly expound a complete theory of evolution that had no holes in it.
Ironically, by fighting teleologists (Creationists), evolutionists were led down the slippery slope toward teleology themselves. This reminds us of the Nietzsche quote, “be careful in fighting monsters that you do not become a monster yourself, for when you gaze into the abyss, the abyss gazes back into you.” An apropos rendition of this quote is: “Be careful when fighting Teleologists, that you do not become a Teleologist yourself; and when you stare long enough into the Peaks and Abysses, the Peaks and Abysses stare back into you.”
We might thus pause and reflect that the obsession with adaptive landscapes is in part what has led evolutionists to unwittingly adopt anti-Darwinian teleological theories of evolution. They have for so long grappled with the meaning of the peaks and abysses of adaptive landscapes, they have not realized that the peaks and abysses have gazed back into them—altering their frame of mind from one that is historical and anti-teleological, to one that invokes final causes (e.g., fitness maximization or gene selection) as explanations for complex traits.
We are now beginning to see also how that a single, dominant worldview on macroevolution takes form. The worldview is one that appeals to selection on long-preserved units like genes, species or clades; or which invokes the final cause of “fitness maximization.” Many evolutionists have unwittingly adopted this singular paradigm, but because of the diversity of ways it manifests, they have been unaware of a shared paradigm.
At this point, however, it is necessary to examine the assumptions that have led evolutionists to adopt this paradigm. A first one is that the only requirement for natural selection is heritable variation in fitness.
To debunk this assumption, it is important to examine the arguments of Lewontin (1970). Lewontin (1970) was the first person to explicitly state the paradigm and attempt a direct refutation of Darwin’s conditions. Lewontin argued that Darwin was incorrect to argue that resource limitation and a struggle for existence are necessary for natural selection. According to Lewontin, the importance of resource limitation and a struggle for existence can be disproved with two arguments.
Lewontin’s first argument is that Darwin’s example of a plant on the edge of the desert struggling against the drought disproves the important of competition among individuals for resources in short supply. However, there is another way to interpret this example. After all, what is drought, except the absence of a necessary resource (water)? The other way to interpret this example is that there are two ways toward resource limitation: the power of population outstripping resources (Malthus’s doctrine), or the environment prematurely restricting resources through drought or famine. Darwin’s example really emphasizes the importance of resource limitation, and another way it can come about (drought or famine, not just power of population).
Lewontin’s second argument is that if two strains of bacteria are increasing exponentially in an excess of nutrients, that the one with a faster division rate will be naturally selected. Thus, according to Lewontin, resource limitation and a struggle for existence is unnecessary for natural selection. However, the other way to interpret this example is as follows. First, one strain will never exclude the other if resources are unlimited. Second, what happens when resources do become limiting? It is quite possible that one strain will feed off the metabolic waste product of the other, or they might somehow otherwise utilize different resources. If so, then the struggle for existence occurs only within strains, and the different strains are best considered separate species. By corollary, if mice and elephants were feeding in excess of food, we should not say that mice are selected relative to elephants simply because they reproduce faster. Once food becomes limiting, the mice and elephants will coexist because they are sufficiently divided in their resource requirements and niche utilization.
Now, there are yet two more requirements to fully abandon this assumption about heritable variation in fitness. First, why did evolutionists uncritically accept Lewontin’s argument? The reason is that they were already convinced. Dominant individuals in the field including Fisher and Dobzhansky disparaged the importance of metaphors like the struggle for existence and the survival of the fittest. As usual, people accepted an assumption as truth because everyone else accepted it.
Second, it is important to recognize just how widely taught this assumption is, and how it influences our worldview. The argument that selection “inevitably follows” from heritable variation in fitness (net reproductive success) is taught in nearly all major evolution textbooks. Moreover, as shown by extracts from textbooks and scientific articles, it influences the way people interpret Darwin and the meaning of his theory. It leads Grafen (2014), for example, to argue that “Darwin (1859) employed evidence of design in the natural world to argue for the existence of natural selection.” But as can be inferred from the previous review of Darwin’s arguments, nothing could be further from the truth. Darwin employed evidence of imperfection as evidence against design.
Now that we have recognized the existence of a single and dominant macroevolutionary paradigm and freed ourselves of the assumption that causes us to unconsciously adhere to this paradigm, we are finally in a position to understand the evidence for the theory of natural reward. We can also feel comfortable comparing this theory to a single teleological paradigm, shared by anyone who extrapolated from natural selection to macroevolution; who employs gene, species or clade selection to macroevolution; or who explains complex traits by appeal to the final cause of “fitness maximization.”
Only three lines of evidence are here discussed. One comes from work on genetic kin recognition, a trait that was formerly explained by teleological appeal to final design. Breaking this trait into simpler components and tracking historical evolution overturns the former teleological theory. It also helps debunk an additional assumption of evolutionists that causes them to uncritically accepted generalized fitness maximization theories. The assumption is that the success of optimization theory in many empirical contexts justifies extrapolation to macro-scales. For example, if optimization theory explains subtle shifts in behavior in well-defined contexts, then it is assumed whole organisms or complex traits are optimizing entities. The example of genetic kin recognition refutes this assumption. This refutation is further supported by discussion of computer simulation studies.
Yet another example comes from sexual reproduction. Work on this topic was constrained because evolutionists did not sufficiently distinguish questions for adaptive maintenance, from questions of macroevolutionary success. Most often, the question asked was why sexual reproduction is widespread, and the answer given is why sexual reproduction is maintained. The answer given is framed in terms of the two-fold cost of males, but it does not actually address the question. In fact, the answer is artificially constrained by the framing of the problem in terms of the two-fold cost. In reality, sexual reproduction can have any number of long-term impacts that have nothing to do with why it is maintained. These can also impact its ubiquity. For example, the effects of sex on hybridization, major genetic change, and spread of transposable elements could all possibly contribute to the ubiquity of sex, even if none of these factors would explain the maintenance of sex.
Finally, and this came up in the Q&A, the theory of natural reward helps us understand the importance of the build up of dispersal morphs build on expanding fronts (as cane toads in Australia). Authors who were constrained by the “fitness” paradigm saw this phenomenon as important for indicating the existence of a process that assembles phenotypes in space rather than time. Shine et al. (2011) took this to be indicative of “natural selection’s shy younger sibling”—spatial sorting—a process that might explain why certain dispersal morphs get fixed on islands. However, when one recognizes the existence of an alternative struggle and evolutionary force, it is possible to view spatial sorting as being indicative of something much more important. When there are open areas for colonization, natural reward favors both those that are more capable of spreading inventions. The differential success of dispersal types on expanding fronts is in fact indicative of natural selection’s bold and entrepreneurial associate, natural reward. In contrast to mere spatial sorting, which can do little more than explain fixation of dispersal types on islands, natural reward is a force of nature that leads to progress and advancement.
As in Darwin’s time, none of these three examples are “smoking gun” evidence for a new theory. However, as more of these examples are assimilated and explained within the umbrella of a new explanatory system, the anomalies of former evolutionary theory will dissolve away (e.g., Crozier’s paradox of genetic kin recognition), there will be greater unification of phenomena (e.g., models of the maintenance versus the success of sexual reproduction), and new and productive lines of inquiry will open up (e.g., on how to use comparative and historical data to determine the sequence of events through which complex traits originate, and the different causes for their success; on the link between microevolutionary and macro- and megaevolutionary processes, and the effects of natural reward over vast time scales).
Finally, Mike Ryan, perhaps playing devil’s advocate, asked why somebody who extrapolates from microevolution to macroevolution would want to adopt this new theory. The first thought that crossed my mind, was simply answering this question with another one. Why would anyone, happy to take design in nature as evidence of a Creator, want to adopt the theory of natural selection? Natural historians in Darwin’s time took the design of organisms as evidence of God’s Divine Providence, just as many today take complex traits as evidence of the power of Natural Selection. Indeed, there is little reason why anyone long accustomed to such ways of thinking would want to abandon them, especially where they justify metaphysical research programs and scientific controversies that have built reputations and careers. That is why I look, as Darwin did, “to young and rising naturalists, who will be able to view both sides of the question with impartiality,” and who will find that they can better “group facts & search out new lines of investigation” on a new theory. New students, you see, are not constrained by their past reputations and do not care if their new discoveries render old areas of research obsolete.
The way I actually answered Mike’s question was, however, quite different. Instead of pointing to the similarity between today’s teleologists and those in Darwin’s time, I focused instead on something I pointed out at the beginning of my “natural reward” paper. Darwin’s mentor and friend, Charles Lyell, had long been a critic of progressionist views. Before 1856, when Darwin told Lyell of his theory, Lyell had long been a critic of naïve progressionist views of evolution, which he saw as being influenced by religious thought. There was nothing in geological history, other than Man coming last, that suggested long-term progressive trends of evolution. Each geological era involved shifts in climates, and different suites of species adapted to those climates. Which species were present was thus context-dependent, not progressive overall.
However, Darwin’s theory suggested that all species had descended from one or a few forms, and this necessitated progressive trends of increasing complexity and diversity. Thus, Darwin’s theory meant that progress must have happened, yet it didn’t really explain this progress. The problem was that natural selection merely adapts species to their current situations, and as Lyell pointed out, a species that had evolved for 100,000 years to tropical and uniform conditions would simply replaced with some other species upon changed conditions. Natural selection thus did not predict progress, and Darwin’s use of terms like “inferior” to refer to earlier species seemed unwarranted. Wallace also recognized this difficulty, and that natural selection has no power to advance any being beyond whatever is necessary to beat its nearest competitors in the struggle for existence.
Both Lyell and Wallace thus remained agnostic to the possibility of the discovery of yet other forces of evolution. They both recognized the importance of natural selection, but did not take it to be the whole truth. As Lyell said in Antiquity of Man p. 365-366,
“Yet we ought by no means to undervalue the importance of the step which will have been made, should it hereafter become the generally received opinion of men of science (as I fully expect it will), that the past changes of the organic world have been brought about by the subordinate agency of such causes as ‘Variation’ and ‘Natural Selection.’ All our advances in the knowledge of Nature have consisted of such steps as these, and we must not be discouraged because greater mysteries remain behind wholly inscrutable to us…”
Wallace (1883) likewise put it well when he stated, “the whole gist of my argument is, not that natural selection ‘does not apply,’ but that it does not exclusively apply, being supplemented by some unknown higher law…” In an earlier work, Wallace (1871) stated, “even if my particular view should not be the true one, the difficulties I have put forward remain, and I think prove, that some more general and more fundamental law underlies that of ‘natural selection…’ It is more probable, that the true law lies too deep for us to discover it; but there seems to me, to be ample indication that such a law does exist…”
It is a somewhat ironical that many evolutionists have accused Lyell and Wallace as being influenced by religion. They were in fact agnostic from a scientific standpoint, while in contrast Darwin religiously extrapolated from a direct struggle for existence to macroevolutionary phenomena. This why Lyell, in refuting Darwin in a letter, was led to the satirical statement that, “I care not for the term ‘Creation’ but I want something higher than ‘selection’, unless the latter divinity (my italics) be supposed to have produced the primeval egg or seed or germ of the first great branches of the organic kingdom—in some ante-Cambrian epoch— To that power I must refer all the wonders of successive groups of species, more specially every step gained in organization & intelligence …” In other words, Lyell attributed to some unknown factor various success advances of organization that are now referred to as evolutionary innovations or “major transitions in evolution.” As I have argued, the force that plays a major role in macroevolutionary advancement is that which rewards innovation through indirect competition during a struggle for supremacy. Natural reward exerts its effects recurrently with the expansion of life, extinction events, and the opening up of new opportunities.
Thus, I have argued that the law of nature, long sought by Darwin in his principle of divergence, and sensed by Lyell and Wallace, is Natural Reward. Darwin’s theory was based on an assumption of resource limitation, which is a transient and local phenomenon. However, Abundance is the dominant condition of nature, and from this stems the Struggle for Supremacy and Indirect Competition manifesting as a Race to Innovate, leading to Natural Reward, The Successive of the Innovative, and major progressive trends of evolution.