Denis Noble Detailed Response: A framework is not a testable theory

Are you ready to discover your college program?

Search Colleges is an advertising-supported site. Featured programs and school search results are for schools that compensate us. This compensation does not influence our school rankings, resource guides, or other information published on this site.

My initial reaction to being paired with David Sloan Wilson in this dialogue was to wonder what we could possibly disagree about. Prof. Wilson has been at the forefront of revealing the limitations and conceptual difficulties of “selfish gene” theory, which many people (including Dawkins himself) actually regard as synonymous with neo-Darwinism.[1] Wilson’s book, with the philosopher Elliot Sober, Unto Others (Harvard UP, 1998), is a landmark publication, widely acknowledged as such both in biology and in economics. The smaller sequel, Does Altruism Exist? (Yale UP, 2015), is also an important book. I did not know of Unto Others when I wrote The Music of Life (Oxford UP, 2006), but I wish I had.

Incidentally, like Wilson, I also believe that research in economics and management theory needs to take account of the new trends in biology. That is why I am one of the organizers of a three-day meeting of the Royal Society and the British Academy on “New Trends in Evolutionary Biology,” which involves leading researchers from the sciences, humanities, and social sciences, and which will be published in the Royal Society journal, Interface Focus. I also lead a relevant project of the Balliol Interdisciplinary Institute, “A Systems Approach to Management Studies,” which is investigating the ways in which the systems approach in biology can inform such approaches in economics and management.

Possible relevance to today’s globalized economy

We are therefore largely on the same wavelength on many of the crucial issues facing both the biological and the social sciences today. These issues are urgent, and pressingly so. It may be far too simplistic to link selfish gene theory to the development of naively reductionist forms of neo-liberalism, and I know that my Oxford colleague Richard Dawkins would certainly not sanction such applications of his theory. But, for good or ill, others have done so. One might therefore go so far as to say that the biological reality or otherwise of altruism is one of the crucial issues of the twenty-first century.

Lehman Brothers PhotoThat is highlighted by the fact that the world is coping with the consequences of catastrophic breakdown in our financial systems, with rapidly growing disparity between rich and poor, and clear evidence of dubiously ethical, even if justified as strictly legal, behavior by far too many people in the top echelons of power in commerce, industry, politics, sport… you name it. The revelations have become so frequent in the media and on the Internet that the world has almost become weary of, and no longer surprised by, the scandalous nature of what emerges. Again, I emphasize that it is a far cry from these revelations to drawing any firm conclusions about the influence of theory in science and economics on people’s behavior. I agree with Professor Wilson’s caution in this regard. But, at least the possibility is there staring us in the face. I referred in my earlier contributions in this dialogue to the influence of some gene-centric interpretations of biology on the eugenics movement in the twentieth century, leading, as we all know, to the horrors of the Second World War, with its undercurrent of the idea of “purifying” the human race. The eugenics movement had precisely this as one of its aims.

In his Statement, Wilson argues very forcefully that

Darwin’s theory did not unleash a plague of toxic social policies justifying inequality. At most, it was added to a quiver that was already full of other arrows, including religious and economic justifications. The claim that Darwin’s theory was used to justify Hitler’s war policies, either directly or indirectly, is demonstrably false…

But the relevant possibility is surely not so much that Darwin’s theories were the key driver, but rather that the later eugenics movement, spearheaded by Francis Galton, Karl Pearson, and Walter Weldon, did provide clear biological arguments for “purifying” the human race. It is significant that it was only after the Second World War that my alma mater, University College London, dropped the title Professor of National Eugenics—a chair that had been founded and personally financed by Francis Galton (below, right). Galton’s chair then became simply the professorship of Human Genetics.

Francis Galton Photo
Sir Francis Galton, FRS (1822–1911)

Moreover, such dangerous interpretations of genetic science have been sufficiently powerful that proponents of the selfish gene view of neo-Darwinism are themselves aware of it and try to counter and distance themselves from such possibilities. As I noted in my Major Statement, Dawkins himself had to write as long ago as 1976, “Let us try to teach generosity and altruism, because we are born selfish.” It is hard to read this statement as anything less than a fear that selfish gene theory might encourage selfish behavior by providing its biological justification.

But what if the assumption that we are “born selfish” simply isn’t justified from a biological viewpoint? That is the conclusion I came to in an analysis of selfish gene theory that I published in 2011.[2] The essence of the article is that:

  1. Selfish gene theory is confused about the definition of a gene in ways that are critical to the distinction between replicators and vehicles that selfish gene theory uses,
  2. Attributing selfishness to a genome sequence is not a testable theory, and
  3. In any case, selfishness is a level-dependent concept. Selfishness at one level does not guarantee selfishness at other levels.

That is why the debate about altruism in general—and this Dialogue in particular—are so important.

I was intrigued, therefore, to see how Professor Wilson would rise to what I thought would be the almost impossible task of defending the self-sufficiency of neo-Darwinism, at least as I and many others understand the term. I have three reasons for being so intrigued.

[Book Image]First, because orthodox neo-Darwinists have systematically refused to engage formally in the discussion. It is 10 years since my book, The Music of Life, was published by Oxford University Press. The challenge to neo-Darwinism was already very clear in that book. That is precisely why so many reviewers of it have been so positive about its message.[3] Yet, not a single orthodox neo-Darwinist has answered it.[4] The same silence has greeted the substantial series of peer-reviewed articles I have published since and which form the basis of The Music of Life Sourcebook. This sourcebook, in turn, has formed the basis of Dance to the Tune of Life, my new book on what I am calling “biological relativity.” That book greatly extends the source material.

My second reason for being intrigued is that self-sufficiency is precisely how the founder of neo-Darwinism, August Weismann, characterized his theory. In a published debate with Herbert Spencer in 1893 he claimed not only that neo-Darwinism was all-sufficient (“Die Allmacht der Naturzüchtung”), but also that it was necessarily so.[5] I now realize through this Dialogue that Weismann’s and Wilson’s uses of the concept of “self-sufficiency” cannot be the same. I will return to this question later because it is the central divide between me and Wilson.

To anticipate the issue briefly, I think that what Wilson and I see as the central features characterizing neo-Darwinism are fundamentally different. Wilson characterizes neo-Darwinism in terms of a necessary research framework—which is based on Tinbergen’s (1963) four questions.[6] By contrast, I characterize neo-Darwinism as the specific and testable hypothesis originally formulated by August Weismann and Alfred Russel Wallace. The first, the research framework, is not falsifiable. It is rather an appeal for a correct methodology in ethology and related fields of science, and I largely agree with the methodology. I say that it is not falsifiable because there is no experiment that could falsify it. It could only be “falsified,” in a rather different sense of that word, if the experience of many researchers using the proposed four question framework showed that it doesn’t or cannot work. It would, I believe, take generations of scientists to reach that conclusion, and I even doubt whether that could ever be established. We can’t know whether a general framework of methodology works or not without trying it, and in science that can take a very long time. It took 30 years, for example, to realize that sequencing genomes would not reveal “The Book of Life.”[7]

By contrast, Weismann’s original theory forming the foundation of neo-Darwinism is not only falsifiable, it has actually been falsified. As I will show in my conclusions, I believe that Professor Wilson and I are actually largely in agreement on these points, though he uses different ways of expressing them.

Tinbergen’s four questions and Wilson’s framework for research

Niko Tinbergen Photo
Nikolaas ‘Niko’ Tinbergen (1907–1988)

I first encountered Tinbergen’s work 50 years ago. In 1966 I was a young academic at Oxford University and had recently won my faculty appointment there on the basis of two 1960 publications in Nature, the first being an experimental discovery of two very different potassium channels in the heart, and the second being a mathematical model to show how this discovery helped to explain electrical activity in the heart and the existence of cardiac rhythm.[8] It was thought, therefore, by the Oxford University science faculty that I was a great rarity in those days: a biologist who could cope with advanced mathematics. Little did they know that, six years earlier, when I struggled to learn enough maths and computer programming to use the vast, but extremely slow, machines called “computers” in those days, I was far from competent in advanced mathematics, or even in quite ordinary mathematics.

I was greatly intrigued by a challenge put to me by the Faculty at Oxford to examine a doctoral thesis of a young research student who was using Laplace transforms to analyse sequence data in ethology experiments done under the supervision of Nikolaas Tinbergen. I didn’t know then, of course, that Tinbergen would, a few years later, win a Nobel Prize in Physiology or Medicine, together with Konrad Lorenz and Karl von Frisch. But I did know of Tinbergen’s great reputation in following Lorenz to bring great rigor to the study of bird behavior. The idea, essentially, was that behavior is an evolvable trait just as much as an organ of the body is an evolvable entity. This was an important insight. The evolution of behavior can be studied scientifically from an evolutionary perspective just as much as the evolution of organs, or any other structure.

Behavior is a process, not a structure, but as I have argued in The Music of Life, and again in Dance to the Tune of Life, even non-behavioral processes in cells, organs, and tissues are what actually evolve generally. Heart rhythm, for example, is a process, not a structure, even though it depends on the microscopic and macroscopic structures of the heart and the organs and systems with which it interacts. The big problem with gene-centric views of biology was to mistake the structure (the DNA sequence) for the process. And that mistake gets hidden by reference to what are called “gene” circuits and “genetic” programs. Those circuits and programs necessarily involve much more than DNA sequences. They could just as readily be characterized as “cell,” “organ,” or “system” circuits or programs. Except that, when we take that into account, the program becomes the functionality. The need for reference to a separate “program” disappears.

Richard Dawkins Photo
Richard Dawkins

This digression explains why I was intrigued by the invitation of the Oxford Faculty in 1966. Here was a doctoral thesis project in ethology supervised by one of the acknowledged founders of the field, in which the student was using mathematics to analyze his results. So, I bought a hefty and expensive tome to bone up on Laplace transforms. I have never regretted that decision. It led 10 years later to the publication of one of my most successful and most mathematical books, still cited frequently today, 40 years on.[9] But I confess that, 50 years later, I don’t remember much of the thesis, other than that it fully deserved the doctorate for which it was submitted. But I can’t forget who the student was, for he became one of the most successful science writers of all time. He was Richard Dawkins—yes, the same person already referred to in this dialogue!—whose writing skills are legendary. I don’t go along with “selfish gene” theory, as I know he knows well, but I greatly admire the skill in communicating his ideas and enthusiasm to the general public. Countless schoolchildren must have been drawn into science through his books. He and I interacted well and courteously in a now legendary debate between him and Lynn Margulis in Oxford in 2009, which I was privileged to chair.

Wilson’s use of Tinbergen’s four questions therefore rings many bells with me. So also does reference to Konrad Lorenz. For it was at the Konrad Lorenz Institute in Vienna in 2013 that I first came across the “extended evolutionary synthesis” (EES), when I was invited by Gerd Müller to lecture there. My lecture used the title “The Music of Life and the billion year dance of the genes,” so the concepts that developed to become Dance to the Tune of Life were already formulated three years ago. Later, when writing one of my recent articles on evolutionary biology, I used the book[10] which emerged from the EES meeting to characterize what I saw as the main differences between the EES and what I have called an “integrative, relativistic” view of evolutionary biology.[11]

Is the “extended evolutionary synthesis” sufficient?

I see the EES as having identified a set of processes in evolution that were not included in either Darwinism or neo-Darwinism. That was a very valuable advance and I have used the book that emerged from the EES meeting many times since. But I have also wondered why the word “extended” was chosen. I am very sympathetic to extensions of theory in science. But I also think that a fundamental rethink is sometimes indicated. I believe that evolutionary biology has reached that stage. The diagrams reproduced below explain the difference between the EES, and my view that, while appreciating the motives and achievements of the EES, we should go beyond it.

EES and neo-Darwinism Diagram
Figure 1. How the EES relates to Darwinism and neo-Darwinism. The developments from Darwinism, to neo-Darwinism, and to the EES are all viewed as extensions of each other. The items in bold font are ones that seem to me to be incompatible with neo-Darwinism as shown in the next diagram (based originally on [10]).

Integrated Synthesis and neo-Darwinism
Figure 2. The relations between Darwinism and later developments, rearranged to show how neo-Darwinism included ideas that have been falsified (bold face). These include the ideas: (1) that DNA alone specifies the organism and ensures its inheritance; (2) that selection operates only on genes as DNA (but that is also where some of the problems lie—see the diagram in my Major Statement explaining the different definitions of a gene); and (3) that the genome is isolated—based on the Weismann Barrier and on the so-called “Central Dogma of Molecular Biology” (based on [11]).

The problem with neo-Darwinism is that it became hardened into a strongly dogmatic insistence that certain processes that predecessors (including notably both Lamarck and Darwin) had proposed to include in evolutionary mechanisms of change were impossible, or even that it was absurd to imagine that they could be included. As I have already shown above, that process began very early in Weismann’s 1893 writings. Essentially, his idea was that blind chance producing genetic variation followed by natural selection to winnow out the failures was not only possible, but also entirely sufficient. Darwin clearly did not believe that.[12] Darwin not only accepted Lamarck’s view that the inheritance of acquired characteristics had played a role, he actually formulated a physical theory (his theory of “gemmules”) for how it could happen that changes in the soma could influence the germline. Recent research has shown that RNAs and imposed genome marking can serve precisely this role.

It seems to me therefore to be one of the original cornerstones of neo-Darwinism to insist that the inheritance of acquired characteristics, via epigenetic and other mechanisms, is impossible. There is now ample experimental evidence that it is possible, and that such transgenerational epigenetic changes can, in some cases, be transmitted down many generations. I dealt with this question in item 7 of my Major Statement. Much more detail and the full references can be found on the relevant pages of my website and in Dance to the Tune of Life (Oxford UP, forthcoming).

Barbara McClintock Photo
Barbara McClintock (1902–1992)

There is also ample evidence that the genome is not isolated. As Barbara McClintock wrote when receiving her Nobel Prize in 1983, the “genome…is a highly sensitive organ of the cell.”[13] James Shapiro has elaborated and extensively documented the evidence for genome re-organization (natural genetic engineering) in response to environmental influences in his book Evolution, a View from the 21st Century.[14]

I conclude, therefore, both that the original formulation of neo-Darwinism was falsifiable and that it has been falsified—if it is regarded, as many neo-Darwinists have insisted, as entirely sufficient.

Of course, that does not mean that the neo-Darwinist mechanism (chance genetic variations followed by natural selection) does not happen. Nor does it mean that all the impressive developments of population genetics using the mathematical ideas developed by neo-Darwinism suddenly become incorrect. That is no more so than a claim that quantum mechanics means that Newton’s equations are no longer useful. It simply means, as Charles Darwin thought, that neo-Darwinism is not the only mechanism by which evolution may have occurred. My appeal is for a return to a more nuanced, multi-process view of evolutionary change. Darwin could not have anticipated what we have discovered today, but in his caution—and even in his specific theory of gemmules—he was far-sighted. If we return to a position that more closely resembles Darwinism by including the inheritance of acquired characteristics, then the term “neo-Darwinism” becomes redundant.

Why the angst?

Why does this seemingly obvious, even innocuous, idea worry so many orthodox neo-Darwinists? So much so that they pour scorn on scientists like me who document the evidence that neo-Darwinism as a scientifically The Creation of Adam Illustratio falsifiable theory is not the only show in town. This is a subject for professional historians and sociologists, of course, but I can at least venture a plausible reason—which is what I do in my new book.

The reason was first pointed out to me many years ago at a 1998 Novartis Foundation Symposium on the Limits of Reductionism in Biology,[15] when a distinguished neuroscientist told me quietly during a coffee break that he would go along with what I was proposing as an integrative view to complement reductionism, “except that it would let God back in.”

It is undeniable that many dogmatic neo-Darwinists are also militant atheists and that they have used their conclusions from their interpretations of evolutionary biology to buttress their position. I am neither a theologian, nor am I conventionally religious. But I also know that many scientists who are conventionally religious and hold theistic views would laugh at some of the caricatures of their views presented by popularizing, atheistic neo-Darwinists. The theologians I know are often better philosophers than their critics and are far removed from the naïve fundamentalist views of many who use their religious beliefs to oppose the idea that life evolved. The problem is that the continuing and often acrimonious debate between neo-Darwinists and their fundamentalist opponents has polarized the debate, particularly so in the United States, to the point at which to back down even marginally on either side would involve great loss of face.

It is in that context that it is possible to understand why it seems necessary to militant atheistic scientists that they should exclude any idea that neo-Darwinism has been shown to be not the only possible mechanism of evolutionary change. That is why some of us who do dissent from the dogmatic forms of neo-Darwinism have formed the Third Way of Evolution website. A steadily growing number of important scientists involved in work on evolutionary biology have joined us. One of my reasons for helping to launch this initiative is that many of the recordings of my own lectures to standard international congresses have been subsequently copied on, or embedded in, websites that are clearly run by creationists or those supporting Intelligent Design theories. I also know that some of my neo-Darwinist critics are quoting from those websites, not from the lectures or publications themselves. They are making the mistake of failing to consult the original sources. Anyone who can write that I proposed that “Darwin was rong” (sic) has simply not even bothered to consult what I said or wrote, particularly because I argue that he was largely right. Sadly, this is typical of the level of debate frequently found on the Internet.

Back to Tinbergen’s four questions

It is therefore a pleasure to take part in a “civil dialogue” with David Sloan Wilson. This is how debate in science should be done and, like him, I am also grateful to for the opportunity.

As I have already noted, Professor Wilson’s definition of neo-Darwinism seems to me to be more like a framework for research—even a necessary framework—than a testable theory of evolution. My reading of Tinbergen’s 1963 article also convinces me that this is the way he viewed his four questions. In his article, he doesn’t point to a way in which his approach could be falsified. Questions, anyway, don’t have the function to be falsified, unless we regard discovering conceptual mistakes embedded in a question to be a form of falsification. For example, a question like “Where is the West Pole?” cannot be interpreted in the same way as “Where is the North Pole?” It suffices to point out that no one knows what a West Pole would be like—or how we would know we had found it—to realize that it is not a valid question. But we know how to find the North and South Poles on a spinning object like the earth. “Where is the West Pole?” might well function in literature or poetry, but it doesn’t have meaning in science. It is simply an ill-formed and useless question.

This short digression on the interpretation of questions is important because it is true to say that Weismann did come to the conclusion that his theory was necessarily true. But a valid empirical theory can’t be necessarily true. Only theories in logic and mathematics can be necessarily true in the sense that they can be proved to be so. Any theory with empirical consequences must run the risk of being falsified if the world turns out not to behave in the way we supposed. I therefore believe that Weismann overstepped himself when he made that claim in his debate with Herbert Spencer. I feel sure that he must have thought he was proposing an empirically valid scientific theory. Why else would he have performed his tail-cutting experiments? Weismann Experiment The problem was that he felt certain he was right and translated that certain belief into his statement about its necessity. We have probably all made that mistake from time to time. It is sufficient to find that we cannot even imagine an exception to our certain belief, to slide into thinking that we have found a necessary empirical truth. Nineteenth-century determinists made a similar mistake when they thought that the world was necessarily determinate and they couldn’t possibly have envisaged the arrival of quantum mechanics.

Tinbergen’s questions seem to me to be a perfectly valid framework for asking the right questions. They are a kind of checklist to tell us whether we really have fully addressed what we need to study about an evolutionary biology problem. As Professor Wilson also says, there is still much research to be done within the framework of those four questions. In that sense, I understand his view that they are sufficient—even that they are necessary.

But I don’t go so far as to regard any other questions as distractions. It does not seem to me to be a distraction to ask whether acquired characteristics can be inherited or whether genomes can be reorganized in response to environmental stress. Those are valid empirical questions and we now have the means for answering them. Moreover, the answers are often interesting and important. They lead to practical conclusions in a variety of fields, ranging from immunology (antigen invasion triggers targeted—and therefore functional—genomic change in immune system cells) to evolutionary biology (the same process occurring intergenerationally via natural genetic engineering) and the growing resistance of bacteria to antibiotics (they behave like the immune system in response to invasive stress).

I will illustrate the statement that acquired characteristics can be inherited with an example that includes an important aspect of animal behavior.

Mouse Maternal Grooming IllustrationThe example concerns a mechanism by which evolution can use epigenetic marking to bypass the Weismann Barrier by transmitting the epigenetic marks to offspring through behavior. This process has been demonstrated by Michael Meany’s group in Canada. Rodents, like many other animals, groom their young by licking and stroking them. This behavior enhances the health and longevity of the progeny. It also influences epigenetic marking in the region of the brain called the hippocampus which, among other roles, plays a part in emotional behavior. The epigenetic effects can therefore predispose the progeny to show the same behavior towards their young. This form of epigenetic inheritance doesn’t even require transmission through the germ line. It is a behavioral way of bypassing the Weismann Barrier.[16]

It seems to me, first, that it was a valid question to ask how this kind of behavior could be transmitted through many generations. In that sense, it was also a question well-formulated within Tinbergen’s framework. If that is what Wilson means by “sufficient,” then I suspect that almost any question about evolution can be interpreted as being within the framework, and it then seems trivially true that we don’t need to go outside the framework. We then arrive at the situation in which one can ask a question within the Tinbergen framework forming Wilson’s definition of neo-Darwinism that leads to a test of a different, and I would argue more usual, definition of neo-Darwinism. For the examples I have given show that acquired characteristics can be robustly inherited, and that in some cases this happens independently of transmission through the germline, contrary to the original formulation of neo-Darwinist theory. There are also examples of epigenetic inheritance that do go through the germline, either as genomic markings, or as accompanying RNAs.[17] These also break the Weismann Barrier.

Now, compare the results of such experiments with Weismann’s statement:[18]

When these deviations only affect the soma, they give rise to temporary non-hereditary variations; but when they occur in the germ-plasm, they are transmitted to the next generation and cause corresponding hereditary variations in the body.

If we take Weismann seriously, Meany’s research clearly disproves his theory, as do a growing number of other examples of the transgenerational inheritance of acquired characteristics that I reference fully in Dance to the Tune of Life. Incidentally, these examples also demonstrate that Weismann’s theory was indeed an empirically testable theory. It was also the foundation stone of neo-Darwinism.

Detailed points using quotations from Wilson’s Major Statement

Finally, before summing up my Response, I would like to tidy up a few detailed points which I find confusing. The first two are quotes from Massimo Pigliucci.

…the [Modern Synthesis] was not something radically different from Darwinism and neo-Darwinism (the late 19th century modification of the original theory that got rid of Lamarckian influences, largely thanks to the work of August Weismann and Alfred Wallace).[19]

This seems to me to be a very strange argument. Neo-Darwinism is significantly different from Darwinism precisely because it “got rid of Lamarckian influences.” That is one of the main reasons I take issue with it. If excluding “Lamarckian influences” means excluding the inheritance of acquired characteristics, then neo-Darwinism was wrong to exclude them—particularly when Darwin himself did not.

This quotation, also from Pigliucci, confirms my view that the exclusion of Lamarckian influences was the main motive and cornerstone of neo-Darwinism:

We are, however, in need of explicitly and organically incorporating into the framework of the MS a number of new discoveries and concepts (phenotypic plasticity, epigenetic inheritance, evolvability, and so forth) that were unknown to, or unappreciated by, the architects of the Modern Synthesis.[20]

The problem here is that epigenetic inheritance is precisely the means by which forms of the inheritance of acquired characteristics become restored to evolutionary biology. So also is phenotypic plasticity, as Waddington showed when he first defined the concept of epigenetics. He also demonstrated ways in which epigenetic changes involving plasticity could become assimilated into genetic inheritance after a few generations. Since the exclusion of the inheritance of acquired characteristics was the main criterion by which Weismann and Wallace distinguished neo-Darwinism from Darwinism, to bring it back is to go back to Darwinism. Neo-Darwinism as defined by Weismann and Wallace is then falsified on its central idea.

It doesn’t matter much whether we talk of “paradigm shifts” or not, but it does matter a lot whether neo-Darwinism remains clearly distinct from Darwinism. If we assimilate precisely what Weismann and Wallace intended to exclude, then I don’t see why we continue to call it “neo-Darwinism.” It reverts to being simply “Darwinism.” This was, after all, why Waddington took issue with neo-Darwinism.

Next, consider this statement by Wilson:

Complex interactions can produce lots of pattern at the system level, but they are no more likely than a point mutation to produce adaptive pattern without a process of selection.[21]

I think this misses an important point about genome reorganization. Shuffling whole domains of sequences, which code for already-functional protein domains, can clearly improve the chances of new functional patterns appearing. To be sure, natural selection can then work on the outcome. But the source of genetic variation is no longer completely blind. If organisms can do this in response to environmental stress, then the process is no more blind than is the response of the immune cells to new antigens. Organisms can recruit stochasticity at the genome level and at the level of gene expression to greatly improve their chances of meeting new challenges.

I also find it hard to reconcile Wilson’s position here with the following later quote:

Anything that smelled of directed evolution was branded with the label “Lamarckian” and declared impossible, much as group selection was declared impossible during the 1960s and 70s.

Backing away from this dogmatic position is an important part of the extended evolutionary synthesis. For me, the best way to think about directed evolution is to focus on animal behavior, which brings the arc of this essay back to Tinbergen. In the conventional view of natural selection, mutations are not directed but they result in the evolution of behaviors that are indubitably directed….

Since the work of James Mark Baldwin, we have known that directed behaviors that are a product of undirected evolution can double back to influence the evolutionary process. What the organism chooses to do by learning alters the selection pressures operating on the genes of. the organism This was celebrated as a major insight at the time—a form of directed evolution that was fully consistent with Weismann’s doctrine….[22]

I like the backing away from “this dogmatic position,” but I am at a loss to reconcile these statements with the earlier ones. It might be best for me to pose my concern as a series of questions, since I may have misunderstood what is intended.

  1. How can “a form of directed evolution” be consistent with “Weismann’s doctrine?” I thought the whole point of Weismann’s work was precisely that evolution is not directed, but rather entirely due to blind chance followed by natural selection.
  2. And how can a form of directed evolution be consistent with the earlier statement that evolution has no foresight? Surely, if we allow evolution to be guided by directed behavior that can “double back to influence the evolutionary process,” then it no longer lacks foresight.
  3. What is meant by saying that “What the organism chooses to do by learning alters the selection pressures operating on the genes of the organism?” If organisms can choose to alter selection pressure, then they are changing the direction of evolution in a functional way.

Next, consider this:

Open forms of phenotypic plasticity are properly regarded as evolutionary processes built by other evolutionary processes, or “Darwin machines” to use the felicitous phrase coined by William Calvin and elaborated upon by Henry Plotkin. Unlike genetic evolution, Darwin Machines are expected to be directed forms of evolution, although they must also have an arbitrary component to remain open-ended. When Darwin Machines become transgenerational through forms of social learning found in many species and forms of symbolic thought that are distinctively human, there should be no stigma whatsoever about the fact that they are partially directed.[23]

Again, I find it hard reconcile this quotation with the statement that evolution “lacks foresight.”

And finally:

In the same way, ideas about directed evolution that have been worked out for the study of behavior are applicable to traits that are not customarily regarded as behavioral, such as genetic and epigenetic inheritance mechanisms, the immune system, developmental programs, and neural processes in the brain. Rigid adherence to Weismann’s doctrine should be declared thoroughly obsolete, along with rigid rejection of group selection. These are profound advances in evolutionary thought, but they do not require going beyond Tinbergen’s four questions. If anything, they require a proliferation of Tinbergen’s four questions for every evolutionary process that is built by another evolutionary process.[24]

I agree that these are “profound advances” and that “Rigid adherence to Weismann’s doctrine should be declared thoroughly obsolete, along with rigid rejection of group selection.” In fact, these are among the primary reasons for which I think that neo-Darwinism as originally formulated has been shown to be insufficient.

By the usual definitions of neo-Darwinism, I think we can safely say that the original theory has been falsified. Moreover, it seems to me that Professor Wilson and I would be in fairly good agreement on the reasons why that is so.

I agree that this doesn’t change anything in relation to Tinbergen’s four questions, but then I also do not think those questions can, in themselves, constitute an empirically testable theory.

Actually, unless I have seriously misunderstood these quotations, my questions on them are rather more than nit-picking. The more I think about them, the more they seem important and the more they seem to support my case.


So where does this leave Professor Wilson and me in this dialogue?

My main conclusion is that Wilson and I seem to agree on many of the new trends in evolutionary biology and their significance, but we are working with fundamentally different definitions of neo-Darwinism. He seems to view neo-Darwinism as a framework rather than an empirically testable hypothesis. In that sense, it is more in the nature of metatheory—a theory about the kinds of questions we should be asking—rather than a hypothesis about the answers to those questions in any particular case.

Metatheory is important and scientists often neglect it. We all need frameworks in our research within which to formulate specific hypotheses. So, I agree that Tinbergen’s four questions are important and that, as questions to guide research, they are sufficient. But I don’t recognise the questions themselves to be a theory of evolution, and therefore not what neo-Darwinism has been about.

August Weismann
August Weismann (1834–1914)

I have encountered some other scientists who believe that neo-Darwinism is more like a framework, and I also believe that this is why they feel they can accept many or even all of the new trends in evolutionary biology without thinking that neo-Darwinism is anything other than all-sufficient (“Allmacht,” to quote Weismann again). But the “necessity” in Weismann’s case cannot be the same as in Wilson’s. Weismann (right) surely believed he had formulated an empirical truth about the world (but mistakenly represented it as a necessary truth). I don’t think Tinbergen’s questions do that. They are not empirical propositions, even though answering them in specific cases could generate empirical statements. But those answers would agree with, or disprove, any hypotheses we may have about those answers, not the general framework questions themselves.

I think we can leave it to the historians of science to work out how neo-Darwinism sometimes seems to morph into metatheory, whereas its founding work was a very strong empirical hypothesis.

Join the Dialogue!

Share your thoughts in the comments section below …while the comments are still open!


1.  It is worth adding that in the last 10 years I have noticed that many evolutionary biologists no longer think that the “selfish gene” represents their case.

2.  D. Noble, “Neo-Darwinism, the Modern Synthesis and selfish genes: are they of use in physiology?,” Journal of Physiology, 2011, 589: 1007–1015.

3.  There are 25 reviews on, with an average score of 4.8. None of them defends neo-Darwinism against the thesis of the book.

4.  Other than through insulting and inaccurate comments on blogsites.

5.  The title of Weismann’s response to Herbert Spencer, Die Allmacht der Naturzüchtung (Jena: Fischer, 1893), means “the all-sufficiency of natural selection.” He wrote: “We accept it [Allmacht]…simply because we must, because it is the only plausible explanation that we can conceive.” He admitted that it was not possible to observe the process in detail, so there could be no experimental proof, but continued “It does not matter whether I am able to do so or not, or whether I could do it well or ill; once it is established that natural selection is the only principle which has to be considered, it necessarily follows that the facts can be correctly explained by natural selection.” For further details on the history of these nineteenth-century debates, see Stephen Jay Gould, The Structure of Evolutionary Theory (Harvard UP, 2002); pp 170–250.

6.  N. Tinbergen, “On aims and methods of ethology,” Zeitschrift für Tierpsychologie, 1963, 20: 410–433.

7.  See Michael Joyner’s lecture, “Chasing Mendel: Thinking about Genotype vs. Phenotype” (University of Oxford, 2016)

8.  This part of my career is described in more detail in the Interview.

9.  J.J.B. Jack, D. Noble, and R.W. Tsien, Electric Current Flow in Excitable Cells. Oxford: Oxford University Press, 1976. This book has been cited nearly 2000 times and is still very influential as the standard text on the mathematics of electrophysiology.

10.  M. Pigliucci, and G.B. Müller, eds., Evolution, The Extended Synthesis. Cambridge, Mass: MIT Press, 2010.

11.  D. Noble, “Evolution beyond neo-Darwinism: a new conceptual framework,” Journal of Experimental Biology, 2015, 218: 7–13.

12.  Darwin wrote in the first edition of On the Origin of Species: “I am convinced that Natural Selection has been the main but not exclusive means of modification” (London: John Murray, 1859; p. 6). He reiterated this statement with increased force in the sixth edition published in 1872.

13.  B. McClintock, “Responses of the Genome to Challenge,” Science, 1984, 226: 792–801.

14.  J.A. Shapiro, Evolution: a view from the 21st century. New York: FT Press, 2011.

15.  Gregory R. Bock and Jamie A. Goode, eds., The Limits of Reductionism in Biology (Novartis Foundation Symposium 213). New York: Wiley, 1998.

16.  I.C.G. Weaver, “Life at the interface between a dynamic environment and a fixed genome,” in D. Janigro, ed., Mammalian Brain Development. New York: Humana Press, 2009; pp. 17–40.

17.  D. Noble, “How widespread is trans-generational inheritance of acquired phenotypic characteristics?” (Music of Life website).

18.  A. Weismann, Das Keimplasma: Eine Theorie der Vererbung. Jena: Fischer, 1892. (English translation)

19.  Massimo Pigliucci and David Sloan Wilson, “The Origin of the Extended Evolutionary Synthesis: An Interview with Massimo Pigliucci” (This View of Life website, 2016).


21.  David Sloan Wilson, Focused Civil Dialogue on Evolution, Major Statement: The Neo-Darwinian Revolution is Far from Complete ( website, 2016; original emphasis).




Take the next step towards your future with online learning.

Discover schools with the programs and courses you’re interested in, and start learning today.

Search Colleges is an advertising-supported site. Featured programs and school search results are for schools that compensate us. This compensation does not influence our school rankings, resource guides, or other information published on this site.
Woman working at desk