Category Archives: Philosophy of Science

Darwinism in the modern era: more on the evolution of evolutionary theory – part II

1 (2)The many conceptual and empirical advances in evolutionary biology during the second half of the twentieth century that I have briefly sketched in part I of this essay naturally led to a broader theoretical turmoil. More and more people felt like the Modern Synthesis (MS) was increasingly becoming too restrictive a view of evolution to keep playing the role of biology’s “standard model.” This group included Carl Schlichting and myself, Mary Jane West-Eberhard (2003), Eva Jablonka, and others. But arguably none made a more concerted, if partial, effort than Stephen Jay Gould in his magnum opus, The Structure of Evolutionary Theory, published in 2002.


The Structure is comprised of two parts, one tracing the history of evolutionary ideas, both pre-and post-Darwin, and the second one presenting Gould’s view of contemporary theoretical debates within the field. While the constructive part of the book focuses too much on paleontology and multilevel selection, Gould correctly identified three conceptual pillars of Darwinism that got imported wholesale into the Modern Synthesis:

1. Agency: the locus of action of natural selection. For Darwin, this was the individual organism, while within the MS the focus expanded to the gene, thus leading to an overall increase of agency. Gould advocated further expansion, to include multiple levels of selection, from the gene to the individual to kin groups to species. This suggestion is perfectly in line with that of other authors advocating an Extended Evolutionary Synthesis (EES).


2. Efficacy: the causal power of natural selection relative to other evolutionary mechanisms. According to Darwin, natural selection is the chief mechanism of evolutionary change, and certainly the only one capable of producing adaptation. The MS formally described—by means of population genetic theory—four additional mechanisms: mutation, recombination, migration, and genetic drift. Gould adds a positive role for developmental constraints to the picture, and advocates of the EES further expand on this theme, including concepts such as those of evolvability (i.e., change over time of evolutionary mechanisms themselves), facilitated variation (from developmental biology), and niche construction (from ecology), among others.


3. Scope: the degree to which natural selection can be extrapolated from micro-to macro-evolutionary outcomes. As we have seen last time, this has been controversial early on, with the MS settling for the same basic picture proposed by Darwin: so-called macro-evolutionary processes are simply micro-evolutionary ones writ large. Gould, of course, questions this, on the basis of the already discussed theory of punctuated equilibria. Proponents of the EES also doubt the received view, suggesting that species selection and group-level ecological characteristics may partially, though not entirely, decouple micro-from macro-evolution.


If Gould’s general take is right, then, evolutionary theory has changed over time and the process can best be tracked conceptually by keeping tabs on changes in the agency, efficacy, and scope of natural selection within the theory. This, incidentally, makes natural selection the fundamental idea in biological evolution, and rightly so. No other concept, not even that of common descent, has had such a complex and convoluted history within the field. Moreover, what the EES is attempting to do can also be understood within Gould’s framework.


Now, as we have seen so far, the latter part of the twentieth century and the beginning of the twenty-first century have seen a renewed debate about the status of contemporary evolutionary theory, with a number of calls for an expansion of the Modern Synthesis into an Extended Evolutionary Synthesis. But what does the latter look like, at the current state of the discussion?


I provided an early sketch of it in a paper published in Evolution back in 2007 (available to Socratic level subscribers from my archives), and an updated and expanded version of that sketch has been put out by Laland and collaborators in 2015. My early analysis began by noting that philosopher Karl Popper famously interpreted the MS as a theory of genes, lacking a comparable theory of forms (i.e., phenotypes). The field got started, however, as a theory of forms in Darwin’s days, with genetics taking on a fundamental role only after the rediscovery of Mendel’s work at the turn of the twentieth century. Consequently, I suggested, a major goal that an EES aims for is an improvement and unification of our theories of genes and of forms. This, seems to me, may best be achieved through an organic grafting of novel concepts onto the foundational structure of the MS, particularly evolvability, phenotypic plasticity (i.e., the ability of a single genotype to produce different phenotypes in response to environmental variation), epigenetic inheritance, complexity theory (from mathematics), and the theory of evolution in highly dimensional adaptive landscapes (from population genetics).


Laland et al.’s paper from 2015 is the most focused and systematic attempt to articulate the EES, explicitly aiming at clearing away inconsistencies in previous works. They begin with a comparison of core assumptions of the MS versus the EES. To give you an idea of what they are getting at, here are the entries for inheritance:


Genetic inheritance (MS): Genes constitute the only general inheritance system. Acquired characters are not inherited.


Inclusive inheritance (EES): Inheritance extends beyond genes to encompass (transgenerational) epigenetic inheritance, physiological inheritance, ecological inheritance, social (behavioural) transmission and cultural inheritance. Acquired characters can play evolutionary roles by biasing phenotypic variants subject to selection, modifying environments and contributing to heritability.


They then run through a series of alternative interpretations of important evolutionary phenomena according to the two frameworks. For instance, in the case of developmental plasticity:


MS: conceptualized as a genetically specified feature of individuals that can evolve under selection and drift. Focus is on the conditions that promote adaptive evolution of plastic versus non-plastic phenotypes. The primary evolutionary role of plasticity is to adjust phenotypes adaptively to variable environments. Plastic responses regarded as pre-filtered by past selection.


EES: considers reducing plasticity to a genetic feature to be explanatorily insufficient. Retains an interest in adaptive evolution of plasticity, but also focuses on how plasticity contributes to the origin of functional variation under genetic or environmental change, and how the mechanisms of plasticity limit or enhance evolvability, and initiate evolutionary responses. Many plastic responses viewed as reliant on open-ended (e.g., exploratory) developmental processes, and hence capable of introducing phenotypic novelty.


Moreover, Laland et al. provide readers with a comparison of different predictions originating from the competing frameworks. For instance, in the case of the relationship between genetic and phenotypic change:


MS: genetic change causes, and logically precedes, phenotypic change, in adaptive evolution.


EES: phenotypic accommodation (a non-genetic process) can precede, rather than follow, genetic change, in adaptive evolution.


Laland et al. also present a graphical outline of the structure of the Extended Evolutionary Synthesis, as they see it . It is instructive to comment on a number of features of their model. Phenotypic evolution—the target of explanation of the entire framework, just as it was for Darwin—is assumed to be affected by three classes of processes: those that generate novel variation, those that bias selection, and those that modify the frequency of heritable variation.


Beginning with the first class, these processes include classical ones like mutation, recombination, gene expression, and developmental regulatory processes. But also EES-specific ones like environmental induction (of developmental processes), niche construction, phenotypic accommodation, and facilitated variation. The second class (processes that bias selection) include only EES-related entries: developmental bias and niche construction, while the third class (processes that affect heritable variation) are all classical (mutation pressure, selection, drift, and gene flow) but are in turn affected by the previous class.


The resulting picture is one of complete and, seems to me, highly coherent, meshing of the MS and the EES perspectives, where the latter adds to but does not really replace any of the previously recognized mechanisms. Which brings me to the next question I wish to address concerning the most recent developments of the now more than 150-year-old Darwinian tradition: is the proposed shift from the MS to the EES akin to a Kunhian paradigm shift?


One of the most controversial aspects of the discussion surrounding the MS versus EES debate is the extent to which the new framework is claimed to be distinct from the old one. At one extreme, there are scientists who simply reject the idea that the EES presents much that is new, claiming that whatever new concepts are being advanced were in fact already part of the MS, either implicitly or explicitly. At the opposite extreme, some supporters of the EES have been making statements to the effect that the new framework somehow amounts to a rejection of fundamental aspects of Darwinism, akin to what philosopher Thomas Kuhn famously termed a “paradigm shift” within the discipline, thus aligning themselves with a tradition that can be fairly characterized as anti-Darwinian. My own position has always been that the truth lies somewhere in the middle (in this case!): the EES is significantly different from the MS, and yet the change does not reflect any kind of scientific revolution within modern biology, but rather more of the same process that has led us from the original Darwinism to neo-Darwinism to the MS itself.


Kuhn famously argued—on the basis, crucially, of examples drawn exclusively from physics—that science goes through an alternation of two phases: during “normal” or “puzzle solving” science, practitioners are focused on addressing specific issues from within a given theoretical framework and set of methods (the “paradigm”), which itself is not the target of empirical testing or conceptual revision. From time to time, however, a sufficient number of “anomalies,” or unresolved puzzles, accumulate and precipitate a crisis within the field. At that point scientists look for a new paradigm, better suited to take into account the insofar unresolved issues. If they find it, the new framework is quickly adopted and deployed in turn to guide a new phase of normal science.


Kuhn suggested a number of approaches to tell whether a paradigm shift has occurred (or, in our case, is in the process of occurring). These include five criteria for theory comparison, as well as three classes of potential incommensurability between theories. Let’s begin by examining the five criteria: (1) accuracy, (2) consistency (internal and with other theories), (3) explanatory scope, (4) simplicity, and (5) fruitfulness of the accompanying research program. Here is how the MS and EES compare, in my mind, according to the Kuhnian criteria:


Accuracy, MS: building on the original Darwinism, it has produced quantitative accounts of the change over time of the genetic makeup of natural populations.


Accuracy, EES: incorporates the same methods and results of both the original Darwinism and the MS, adding the explanation of developmental and other self organizing biological phenomena.


Consistency, MS: as internally consistent as any major scientific theory, features explicit external links to genetics, molecular biology, and ecology.


Consistency, EES: same degree of internal and external consistency as the MS, with the addition of external links to developmental biology, genomics, and complexity theory, among others.


Scope, MS: new facts about the biological world that are explained have been consistently uncovered for the past several decades.


Scope, EES: further expands the scope of the MS by explicitly including questions about the origin of evolutionary novelties, the generation of biological form, and the problem of genotype–phenotype mapping.


Simplicity, MS: uses a limited number of mechanisms (natural selection, genetic drift, mutation, migration, assortative mating) to account for evolutionary change over time.


Simplicity, EES: makes use of all the mechanisms of the MS, adding a number of others such as epigenetic inheritance, evolvability, facilitated (i.e., self-emergent) variation, etc.


Fruitfulness, MS: has a history of more than 70 years of vigorous research programs, building on the previous fruits of the original Darwinism.


Fruitfulness, EES: builds on the ongoing research program of the MS but has also already led to empirical (e.g., emergent properties of gene networks and of cell assemblages) and conceptual (e.g., evolvability, phenotypic plasticity) discoveries, though of course it is very much a work in progress as of the moment of this writing.


Even this brief survey ought to make it clear that the MS => EES is not a paradigm shift, but rather an organic expansion. Then there is the second test proposed by Kuhn to consider, a test in a sense more stringent, that of incommensurability. If two theories are incommensurable in even one of the three classes, a good argument can be made that a paradigm shift is occurring. The classes in question are methodological, observational, and semantic.


Methodological incommensurability refers to the notion that different paradigms lead scientists to pick different “puzzles” as objects of research, as well as to the idea that scientists then develop distinct approaches to the solution of those puzzles. The EES takes on board the same puzzles, and the same set of approaches, of the MS, but it also adds new puzzles (such as the appearance of so-called evolutionary novelties, like eyes, feathers, spines, and so forth), which were largely untouched, or dealt with only superficially, by the MS. It further adds new approaches, like interpretations of evolutionary changes in terms of niche construction, developmental plasticity, or epigenetic inheritance.


Observational incommensurability is tightly linked to the idea that observations are theory dependent: what is considered a “fact” within one theoretical context may not be such in a different theoretical context. For instance, in pre-relativity physics there was a (supposed) fact of the matter that some kind of substance, referred to as ether, had to be present in space in order for light to travel through it. After the famous Michelson–Morley experiment demonstrating that there was no such thing as ether, the relevant fact became the constancy of the speed of light and therefore the relativity of frames of reference. Nothing like that seems to be happening in evolutionary biology at the moment: the very same facts that have been catalogued and explained by the MS enter into the empirical corpus of the EES, to be further expanded with new facts that come to the forefront because of the additional conceptual advancements.


Semantic incommensurability has to do with shifts in the meaning of terms used by scientists, one of Kuhn’s examples being that of “mass,” which is a conserved, static quantity in Newtonian mechanics, but becomes interchangeable with energy within the framework of Einstein’s relativity. Again, I do not discern any analogous shift in the terminology used by proponents of the MS versus EES. Key biological concepts, such as species, genes, phenotypes, niche, and so forth, retain similar and perfectly commensurable meanings, even though our understanding of their referents becomes increasingly sharp.


It seems, therefore, that Darwinism after the Modern Synthesis has proceeded along similar lines to those followed by Darwinism before the MS: a continuous expansion of both empirical knowledge and conceptual understanding, an expansion that is likely to continue for the remainder of the current century and beyond.


This discussion is in part an opportunity to call for a bit of house cleaning, so to speak, on the part of evolutionary biologists and philosophers of science. For instance, it is truly astounding that in France the Modern Synthesis, and in particular population genetics, was not included in standardized university curricula, or addressed within main research programs until the 1970s. Against the Darwinian picture that was developing abroad, French life scientists supported various forms of Lamarckism throughout the twentieth century, and some of that attitude still lingers. There is no good scientific reason for that, and it is hard not to pin such an attitude on sheer nationalism and the cultural worship of Lamarck. Needless to say, that sort of thing has no place in a mature science. The French are not the only culprits here, and the fact that there are “German,” “Russian,” and other “traditions” within evolutionary biology is more than a little bizarre.


It’s also somewhat surprising that behavioral biologists are still clinging to simplistic notions from sociobiology and evolutionary biology, which have long since been debunked. It’s not the basic idea that behaviors, and especially human behaviors, evolve by natural selection and other means that is problematic. The problem, rather, lies with some of the specific claims made, and methods used, by evolutionary psychologists.


It is also both surprising and problematic that some researchers are still pursuing non-“mechanistic” or non-“physicalist” research programs, whatever that means. Indeed, a major point of the EES is to help bring the focus back on the organism and even the ecosystem, and yet—as I just argued above—this does not require a wholly alternative synthesis at all.


Over time, Darwinism has advanced its own agenda by incorporating a variety of themes proposed by its critics, including “saltationism” (punctuated equilibrium) and “Lamarckism” (epigenetic inheritance, phenotypic plasticity, and niche construction). This is fine, so long as we keep in mind that the terms within scare quotes above are to be understood in a modern, radically updated sense, and not along the lines of what biologists were thinking decades or even centuries ago. It’s this inherent flexibility of Darwinism that has allowed people with views as divergent as Stephen Jay Gould and Richard Dawkins to (rightly) claim the Darwinian mantle.


This ability to incorporate critical ideas is neither just a rhetorical move nor somehow indicative of serious problems inherent in the Darwinian approach. In the end, the various Darwinian traditions in evolutionary biology are best understood as a wide ranging family of conceptual and research approaches, always in dialectic dialogue with each other, always in a constructive tension that transcends the agendas and (sometimes strong) personalities of the many individual scientists that recognize themselves as intellectual descendants of Charles Darwin. More than a century and a half later, evolutionary theory keeps evolving.

Darwinism in the modern era: more on the evolution of evolutionary theory – part I

1 (1)Scientific theories are always provisional accounts of how the world works, intrinsically incomplete, and expected to be replaced by better accounts as science progresses. The theory of evolution, colloquially referred to as “Darwinism,” is, of course, no exception. It began in 1858 with joint papers presented to the Linnaean Society by Charles Darwin and Alfred Russell Wallace and was formalized shortly thereafter in On the Origin of Species. The original theory featured two conceptual pillars: the idea of common descent (which was accepted by a number of scholars even before Darwin), and that of natural selection as the chief mechanism of evolution, and the only one capable of generating adaptation.


The first bit of tinkering took place shortly thereafter, when Wallace himself, together with August Weismann, proposed to drop any reference to Lamarckian theories of heredity because of the newly proposed notion of the separation between sexual and somatic cellular lines, thus generating what is properly known as neo-Darwinism. After undergoing a temporary crisis, as a result of increasing skepticism from paleontologists and developmental biologists, we enter two phases of the so-called Modern Synthesis, the biological equivalent of the Standard Model in physics: the first phase consisted in the reconciliation between Mendelism (i.e., genetics) and Darwinism (i.e., the theory of natural selection), leading to the birth of population genetics; the second phase consisted in an expansion of the theory to include fields like natural history, population biology, paleontology, and botany.


What happened to “Darwinism” after 1950? The Modern Synthesis (MS) reigned as the dominant paradigm in the field, rather unchallenged until the late 1980s and early 1990s. At which point a number of authors, coming from a variety of disciplines, began to question not so much the foundations but the accepted structure of the MS. By the very late twentieth-century and early twenty-first-century, calls to replace the MS with an Extended Evolutionary Synthesis (EES) had begun to grow loud, and to be countered by equally loud voices raised in defense of the MS. How did this happen, and what does it mean for the current status and future of evolutionary theory? To understand this we need to step back for a moment and take a broad view of conceptual developments in the biological sciences during the second half of the twentieth century.


The second half of the twentieth century has been an incredibly exciting time for biology, a period that has put the discipline on the map at least at the same level of interest as physics, the alleged queen of sciences, and arguably even more so. Let me remind you of some of the major developments that have made this possible, because they all—directly or indirectly—eventually fed into the current discussion over the MS versus the EES as dominant conceptual frameworks in evolutionary biology.


A major breakthrough in one of the foundational fields of the Modern Synthesis, population genetics, came with the invention of a technique called gel electrophoresis, which for the first time made it possible to directly assess protein and gene frequencies in large samples drawn from natural populations. While research on electrophoresis began as early as the 1930s, it was the breakthrough work of Richard Lewontin and John Hubby in 1966 that set population genetics on fire. The unexpected discovery was, as the authors put it, that “there is a considerable amount of genic variation segregating in all of the populations studied …[it is not] clear what balance of forces is responsible for the genetic variation observed, but [it is] clear the kind and amount of variation at the genic level that we need to explain.” This new problem posed by a much larger degree of genetic variation than expected in natural populations eventually led to a revolution in population genetics, and also directly to the origination of the impactful neutral theory of molecular evolution first proposed in 1968 by Motoo Kimura.


The neutral theory was a landmark conceptual development because for the first time since Darwin it challenged the primacy of natural selection as an agent of evolutionary change. To be sure, Kimura and colleagues didn’t think that phenotypic evolution (i.e., the evolution of complex traits, like eyes, hearts, etc.) occurred in a largely neutral fashion, but if it turned out that much of what goes on at the molecular level is independent of selective processes, then the obvious question is how is it possible that largely neutral molecular variation can give rise to non-neutral phenotypic outcomes. Eventually, the debate about the neutral theory—which raged on intensely for a number of years—was settled with a sensible and empirically consistent compromise: a lot of molecular variation is “near-neutral,” which means that the role of stochastic processes such as genetic drift at the molecular level is significantly higher than might have been expected on the basis of a face-value reading of the tenets of the Modern Synthesis.


What could possibly connect the near-neutral molecular level with the obviously functional and therefore likely selected phenotypic level? The obvious answer was: development. The only problem was that developmental biology had famously been left out of the Modern Synthesis. It looked like something was seriously amiss with modern evolutionary theory.


Things began to change as an offshoot of yet another revolution in biology: the rapid advances made in molecular biology after the discovery of the structure of DNA in 1953. While molecular biology kept accelerating its pace independently of organismal biology for several decades—until their confluence in the era of evolutionary genomics—in the late 1970s the existence of homeotic genes regulating embryonic patterns of development in Drosophila was discovered. It soon turned out that this and similar classes of regulatory genes are both widespread and evolutionarily conserved (i.e., they don’t change much over time), so that they are one of the major keys to the understanding of the complex interplay among genotype, development, and phenotype.


This new approach eventually flourished into a new field, known as evolutionary developmental biology, or evo-devo for short, and one of its major contributions so far has been a marked shift of emphasis in the study of morphology and development, from the sort of classical population genetic studies focused on structural genes to an emphasis on regulatory genes and their potential to help us build a credible theory of the origin of evolutionary novelties (i.e., new structures like wings or flower). As Prud’homme and colleagues put it in 2007:


Because most animals share a conserved repertoire of body-building and -patterning genes, morphological diversity appears to evolve primarily through changes in the deployment of these genes during development. … Morphological evolution relies predominantly on changes in the architecture of gene regulatory networks and in particular on functional changes within [individual] regulatory elements. … Regulatory evolution: (i) uses available genetic components in the form of preexisting and active transcription factors and regulatory elements to generate novelty; (ii) minimizes the penalty to overall fitness by introducing discrete changes in gene expression; and (iii) allows interactions to arise among any transcription factor and [regulatory genes].


The picture that emerges from this and many other studies is not incompatible with the simple mathematical models that were incorporated into the Modern Synthesis, but it does present us with a much more complex and nuanced understanding of genetic, developmental, and phenotypic evolution, so much so that it is little wonder that people have been increasingly referring to the current, very much in flux, version of evolutionary theory as the Extended Synthesis.


I have already mentioned the molecular biology revolution initiated in the 1950s, which eventually led to the genomic revolution. Both these radical developments initially affected evolutionary biology only indirectly, by providing increasingly powerful new analytical tools, such as gel electrophoresis, and later on gene sequencing. But inevitably genomics itself became an evolutionary science, once technical developments made it possible to sequence entire genomes more quickly and cheaply, and molecular biologists fully internalized, as Theodosius Dobzhansky famously put it, that nothing in biology makes sense except in the light of evolution. The structure and function, as well as the sheer diversity, of genomes are themselves not understandable if not through evolutionary lenses, so that genomics and evolutionary biology currently represent a rare example of synergism between scientific disciplines: the first provides tools for the latter to advance, while the second one allows for a theoretical understanding of the data that the first one accumulates at such a heady pace.


While of course other disciplines within biology have made progress during the second part of the twentieth century—ecology, for instance—the next bit of this panoramic view I wish to briefly comment on concerns yet another area of inquiry that had played only a secondary role during the Modern Synthesis: paleontology. The field had always been a thorn in the side of Darwinism, since many paleontologists early on had rejected the Darwinian insight, proposing instead the idea that macro-evolutionary change was qualitatively distinct from the sort of micro-evolution that Darwin famously modeled on the basis of plant and animal breeding (and of course, notoriously, creationists have always made a big deal of the distinction between micro- and macro-evolution, often without understanding it). Indeed, it was this very rejection, together with the apparent incompatibility of Mendelism and Darwinism, that led to the above mentioned period of “eclipse” of the Darwinian theory at the turn of the twentieth century.


Paleontology’s early alternative to Darwinism took the shape of orthogenetic theory (according to organisms change in the same direction over millions of years), which in turn was essentially a scaled-up version of Lamarckism, since it postulated an inner vital force responsible for long-term evolutionary trends, which many paleontologists saw as otherwise inexplicable within the Darwinian framework. It was George Gaylor Simpson’s magistral role within the Modern Synthesis that cleared away any remnants of orthogenesis from paleontology, doing for that field what Fisher, Haldane and Sewall Wright had done for Mendelian genetics: he convincingly argued that the sort of so-called “micro”-evolutionary processes accounted for by Darwinism could be extrapolated to geological timescales, thus yielding the appearance of macro-evolutionary changes of a qualitatively different nature. In reality, Simpson argued, the second is simply a scaled up version of the former.


Simpson, however, was arguably too successful, essentially making paleontology a second-rate handmaiden to population genetics while overlooking the potential for its original contributions—theoretical as well as empirical—to the overall structure of evolutionary theory. Eventually, Simpson’s “conservatism,” so to speak, led to a backlash: Niles Eldredge and Stephen Jay Gould, the enfants terribles of modern paleontology, published in 1972 a landmark paper proposing the theory of punctuated equilibria, according to which evolution, when seen at the macroscopic scale, works by fits and starts: long periods of stasis during which not much appears to be happening in a given lineage, interrupted by sudden “bursts” of phenotypic change. The theory was immediately misunderstood by many population geneticists, who thought that Eldredge and Gould were attempting to revive an old notion known as “hopeful monsters,” i.e., of instantaneous evolutionary change resulting from genome-wide restructuring.


To be fair, at some point Gould’s own anti-establishment rhetoric, and the fact that creationists often mentioned him in their support, contributed to the confusion. But in fact, the sort of punctuations that Eldredge and Gould saw in the fossil record takes place over tens of thousands of generations, thus leaving plenty of time for standard Darwinian processes to do their work. As they pointed out later on in the debate, the real novel issue is that of prolonged stasis, over millions of years, not the allegedly (but not really) “instantaneous” change. A major class of explanation proposed especially by Gould for this observed stasis had to do with developmental processes and constraints, which nicely connects the new paleontology with the emerging field of evo-devo mentioned above, making both of them into pillars of the ensuing Extended Synthesis in evolutionary biology.


(next time: the Stephen Jay Gould conceptual revolution and the birth of the Extended Evolutionary Synthesis)

The impossible conundrum: science as a (perennial?) candle in the dark

1(left: Carl Sagan; right: Richard Lewontin)


When I was a kid I wanted to be an astronomer. One of my role models was Carl Sagan, the charming original host of the television series Cosmos and author of countless books on astronomy and the nature of science. Later on I decided that biology was really my calling, and my entire career was the result of reading a single, incredibly powerful paper: The analysis of variance and the analysis of causes, by Richard Lewontin. I never had the pleasure of meeting Sagan, but I did have an hour long chat with Lewontin when I was a graduate student at the University of Connecticut and he was visiting our lab. It was one of the highlights of my life.


Both Sagan and Lewontin had far more impact on me than just their science. Sagan made me sensitive to the importance of communicating with a broader public, to share the wonders of the scientific worldview, as well as to fight the irrationality of pseudoscience. Lewontin made me sensitive to the ideological underpinnings of science and even science popularizing, and therefore, ironically, somewhat skeptical of Sagan’s own approach.


Recently, one of my readers suggested that I take a fresh look at a classic within this context: Lewontin’s review of one of Sagan’s best known books, and one that has influenced me for two decades: The Demon-Haunted World, subtitled Science as a Candle in the Dark. The review, entitled Billions and Billions of Demons (a playful, perhaps somewhat sarcastic, take on Sagan’s famous tagline about a universe with billions and billions of stars) is well worth pondering again today.


Lewontin opens with a recounting of when he met Sagan for the first time, on the occasion of a public debate about creationism vs evolution in Little Rock, Arkansas, in 1964. The experience was formative for both, but they came away from it with radically different messages:


“Sagan and I drew different conclusions from our experience. For me the confrontation between creationism and the science of evolution was an example of historical, regional, and class differences in culture that could only be understood in the context of American social history. For Carl it was a struggle between ignorance and knowledge.”


I can sympathize. When, in 1997, I first debated a creationist, Duane Gish of the Institute for Creation Research (no kidding), I was squarely looking at things through Sagan’s filter: obviously creation “science” is no such thing; obviously evolutionary theory is solid science; and obviously anyone disagreeing with these two propositions is a hillbilly ignoramus. More than two decades after that debate I think that position was incredibly naive, and I find myself far closer to Lewontin’s, though not entirely on board just yet.


As Lewontin aptly puts it:


“The primary problem is not to provide the public with the knowledge of how far it is to the nearest star and what genes are made of, for that vast project is, in its entirety, hopeless. Rather, the problem is to get them to reject irrational and supernatural explanations of the world, the demons that exist only in their imaginations, and to accept a social and intellectual apparatus, Science, as the only begetter of truth. The reason that people do not have a correct view of nature is not that they are ignorant of this or that fact about the material world, but that they look to the wrong sources in their attempt to understand.”


In other words, and contra Sagan, it isn’t a question of educating people about facts, it’s a question of convincing them to trust the better authority. Think of it this way. You probably “know” that atomic nuclei are made of quarks, right? But do you? Really? Unless you are a physicist, or at any rate someone whose grasp of physics is far better than average, you don’t actually know how science arrived at this basic fact about the structure of the world. Instead, you are simply repeating a statement that you read in a book or heard from a prominent physicist, or your college physics professor. You don’t know. You trust.


That’s why rejection of evolution in favor of creationism — while wrong (I actually know this, I’m a biologist) — is not irrational. It simply means that many people in the United States would rather trust their preachers, who they think speak on behalf of God, than Profs. Sagan, Lewontin, or Pigliucci. That’s why Lewontin, correctly, says that the only way to understand why creationism is such an issue in the US of A but not in pretty much any other Western country (and, again, is very much an issue in a lot of Islamic countries) we don’t need to look at the quality of science education. We need to look at the specific cultural history of the United States vs that of European countries.


Sagan did not get it. Here is Lewontin again:


“The only explanation that [Sagan] offers for the dogged resistance of the masses to the obvious virtues of the scientific way of knowing is that ‘through indifference, inattention, incompetence, or fear of skepticism, we discourage children from science.’ He does not tell us how he used the scientific method to discover the ‘embedded’ human proclivity for science, or the cause of its frustration. Perhaps we ought to add to the menu of Saganic demonology, just after spoon-bending, ten-second seat-of-the-pants explanations of social realities.”


You hear similar ex cathedra pronouncements from the contemporary heirs of Sagan’s approach, for instance Neil deGrasse Tyson (who has taken over the helm of the new Cosmos series). Their analysis of the hows and whys of widespread beliefs in parapsychology, UFOs, astrology and so forth is just as unempirical and “seat-of-the-pants” as Sagan’s. One would expect better from people who loudly insist on the absolute necessity of systematic empirical data before making any pronouncement.


Lewontin then proceeds with chastising another common Sagan-Tyson-et-al argument in defense of science: that it “delivers the goods.” Well, yes, sometimes. At times, though, those “goods” are anything but (atomic weapons, biological weapons, Facebook), and in other cases there is no delivery at all (the “war on cancer,” or the over-hyped promises of the human genome project). Meanwhile billions and billions — of dollars — are spent at taxpayers’ expense. Referring to the repeated promises of scientists to deliver cures for diseases if they were only given money to sequence the genes associated with them, followed by inevitable failure since a DNA sequence by itself doesn’t provide a cure for anything, Lewontin writes:


“Scientists apparently do not realize that the repeated promises of benefits yet to come, with no likelihood that those promises will be fulfilled, can only produce a widespread cynicism about the claims for the scientific method. Sagan, trying to explain the success of Carlos, a telepathic charlatan, muses on ‘how little it takes to tamper with our beliefs, how readily we are led, how easy it is to fool the public when people are lonely and starved for something to believe in.’
Not to mention when they are sick and dying.”


Ouch, but on the mark. And there is more where that came from:


“Sagan’s suggestion that only demonologists engage in ‘special pleading, often to rescue a proposition in deep rhetorical trouble,’ is certainly not one that accords with my reading of the scientific literature. … As to assertions without adequate evidence, the literature of science is filled with them, especially the literature of popular science writing.”


I must say that my own experience as a scientist first, and now as a philosopher of science, is far more in synch with Lewontin’s cynicism than with Sagan’s optimism.


And here is another gem from the review:


“When, at the time of the moon landing, a woman in rural Texas was interviewed about the event, she very sensibly refused to believe that the television pictures she had seen had come all the way from the moon, on the grounds that with her antenna she couldn’t even get Dallas. What seems absurd depends on one’s prejudice. Carl Sagan accepts, as I do, the duality of light, which is at the same time wave and particle, but he thinks that the consubstantiality of Father, Son, and Holy Ghost puts the mystery of the Holy Trinity ‘in deep trouble.’ Two’s company, but three’s a crowd.”


Just in case your blood is boiling and you begin to think Lewontin to be a postmodern deconstructionist, think again (and try to breathe deeply). He is an atheist, and he certainly does believe that we landed on the moon. His point is about cautioning scientists and science popularizers against dismissing others on the ground that their views are “obviously” irrational. Rationality is a great tool, but its deployment depends on one’s axioms or, as Lewontin’s puts it, one’s prejudices.


Here is where I partially, but only partially, part company with Lewontin:


“We take the side of science in spite of the patent absurdity of some of its constructs, in spite of its failure to fulfill many of its extravagant promises of health and life, in spite of the tolerance of the scientific community for unsubstantiated just-so stories, because we have a prior commitment, a commitment to materialism.”


Well, yes, sort of. I would say that materialism itself is a philosophical position that many have arrived at because it is the one that makes the most sense of the world as we understand it. But wait, isn’t our understanding of the world based on the assumption of materialism? In a sense, but I think it is a mistake to see one as definitely preceding the other. Materialism and science co-evolved for centuries, and there was plenty of time when many prominent scientists were definitely not materialists, or at least not thoroughgoing materialists — from Newton to Alfred Wallace (the co-discoverer of natural selection). But the more the metaphysical leanings of natural philosophers (as scientists were once called) approached full fledged materialism, the more their science became successful at explaining and manipulating the world. This is, in a sense, a beautiful, centuries-long example of why one’s metaphysics should never be far from one’s epistemology (as it is, by contrast, with religion). The problem is that it’s really hard to imagine how to trigger that same sort of shift in a general public that hardly thinks either philosophically or scientifically. And no, more courses along the lines of Biology or Physics 101 ain’t gonna do it.


Lewontin, again, is far more perceptive than Sagan:


“The struggle for possession of public consciousness between material and mystical explanations of the world is one aspect of the history of the confrontation between elite culture and popular culture. … Evolution, for example, was not part of the regular biology curriculum when I was a student in 1946 in the New York City high schools, nor was it discussed in school textbooks. In consequence there was no organized creationist movement. Then, in the late 1950s, a national project was begun to bring school science curricula up to date. … The elite culture was now extending its domination by attacking the control that families had maintained over the ideological formation of their children. The result was a fundamentalist revolt, the invention of ‘Creation Science,’ and successful popular pressure on local school boards and state textbook purchasing agencies to revise subversive curricula and boycott blasphemous textbooks.”


Lewontin is absolutely right here. But the problem is, and he would be the first one to admit it, that there is no solution in sight. Are we supposed not to teach one of the most important scientific theories of all time because teaching it is going to be taken as yet another affront perpetrated on the working class by the moneyed elite? I doubt it. But the only other path I can see just ain’t gonna happen: establish a society where there is no such thing as the moneyed elite, where everyone has access to free education, and where consequently a lot of the cultural and economic factors that Lewontin correctly pinpoints will be erased or at least greatly diminished. I’ not holding my breath, are you?


The review concludes with a quote from the Gorgias, one of Plato’s dialogues (which Sagan would have appreciated, though I’m pretty confident that a lot of contemporary science popularizers have no idea why anyone would quote a philosopher who’s been dead more than two millennia. After all, isn’t philosophy useless?). Gorgias, a sophist, and Socrates are debating the relative virtues of rhetoric and technical expertise in public life. We are meant, of course, to sympathize with Socrates, but see if you can appreciate Gorgias’ point, in light of the preceding discussion:


Gorgias: “I mean [by the art of rhetoric] the ability to convince by means of speech a jury in a court of justice, members of the Council in their Chamber, voters at a meeting of the Assembly, and any other gathering of citizens, whatever it may be.”


Socrates: “When the citizens hold a meeting to appoint medical officers or shipbuilders or any other professional class of person, surely it won’t be the orator who advises them then. Obviously in every such election the choice ought to fall on the most expert.”


Obviously it ought, but equally obviously it doesn’t. And that, two and a half millennia later, is still the problem, and the reason why we are in the mess we are in.

No, science does not provide all the answers to the big questions

From time to time a famous scientist allows himself (in my experience it’s always a man) to write nonchalantly about something of which he demonstrably has only a superficial grasp: philosophy. The list of offenders is a long one, and it includes Lawrence Krauss, Neil deGrasse Tyson, and Stephen Hawking, among several others. (Fortunately, there are also exceptions, scientists who value a constructive intercourse with the humanities, like Sean Carroll.) The latest entry in this dubious pantheon is Peter Atkins, who recently published a sloppy essay in the otherwise excellent Aeon magazine entitled “Why it’s only science that can answer all the big questions.” Oh boy.


Atkins begins by telling us that there are two fundamental kinds of “big questions”:


“One class consists of invented questions that are often based on unwarranted extrapolations of human experience. They typically include questions of purpose and worries about the annihilation of the self, such as Why are we here? and What are the attributes of the soul? They are not real questions, because they are not based on evidence. … Most questions of this class are a waste of time; and because they are not open to rational discourse, at worst they are resolved only by resort to the sword, the bomb or the flame. … The second class of big questions concerns features of the Universe for which there is evidence other than wish-fulfilling speculation and the stimulation provided by the study of sacred texts. … These are all real big questions and, in my view, are open to scientific elucidation.”


This is not news, of course, at all. David Hume — one of my favorite philosophers — made essentially the same argument back in the 18th century, in his case rejecting what he saw as the waste of time associated with the Scholastic metaphysics that had prevailed throughout the Middle Ages:


“If we take in our hand any volume; of divinity or school metaphysics, for instance; let us ask, Does it contain any abstract reasoning concerning quantity or number? No. Does it contain any experimental reasoning concerning matter of fact and existence? No. Commit it then to the flames: for it can contain nothing but sophistry and illusion.” (An Enquiry Concerning Human Understanding)


With all due respect to Hume, it’s a good thing people didn’t follow his advice, or we would have lost his very own Enquiry Concerning Human Understanding, since that book doesn’t contain any abstract reasoning concerning quantity or number, nor does it contain any experimental reasoning concerning matter of fact. And yet, it is — justly — considered to be one of the most important works of modern philosophy.


Atkins apparently realizes that he may come across as a bit too facile, since he acknowledges that he is defining the big questions precisely as those that science can answer, turning then around to “demonstrate” that science is the only discipline equipped to answer such questions. As he drily puts it when considering the obvious charge of circularity: “that might be so.” Which doesn’t stop him from proceeding as if it were not so.


Atkins tells us that science is getting ready to tackle what he considers the next three big questions: How did the Universe begin? How did matter in the Universe become alive? and How did living matter become self-conscious?


I have no doubt, as a scientist, that those are, indeed, scientific questions. I’m slightly more skeptical, as a philosopher, that science will actually be able to come up with answers. Fundamental physics, after more than a century of uninterrupted success, seems to have entered a period of navel gazing where speculation (admittedly mathematically informed speculation) is poised to replace empirical evidence. So we shall see if and when we’ll actually get a “theory of everything,” and whether that theory will in fact be able to tell us how the universe began from “nothing” (there is some doubt that it will).


Regarding the second question, the origin of life, theories have been piling up for several centuries now, and yet we don’t seem to be particularly close to a resolution just yet. I’m certainly not arguing that it isn’t possible, but it’s a very, very difficult problem, for the simple reason that a lot of the historical traces have been lost. No geological strata survive from the time when the primordial earth was home to the first living organisms, meaning that researchers on the origin of life are like detectives who already know the smoking gun isn’t going to be found. At best, they’ll have to rely on circumstantial evidence. Even should we be able to produce life artificially in the laboratory that would not solve the problem, since it wouldn’t mean that life on our planet actually followed anything like that particular causal path.


As for consciousness, I remain convinced that the problem is indeed biological in nature, and that therefore developmental, evolutionary, and neuro-biology are the disciplines best positioned to find a solution. But at the moment nobody seems to have much of a clue, and common talk of the brain being a computer is finally beginning to be understood as the shaky and very likely misleading analogy that is.


So, yes, if any of those three big questions are going to be answered, the answer will be a scientific one. But what about other questions that arguably just as big (or, for most of us, even bigger)? Here Atkins shifts into full scientistic mode:


“I see no reason why the scientific method cannot be used to answer, or at least illuminate, Socrates’ question ‘How should we live?’ by appealing to those currently semi-sciences (the social sciences) including anthropology, ethology, psychology and economics.”


Please notice a number of interesting and revealing things about this sentence. First, Atkins is making the time-honored argument from personal incredulity: “I see no reason why…” Which, of course, is not an argument at all, but an elementary logical fallacy. Second, he is seriously hedging his bets when he immediately qualifies his initial statement: “or at least illuminate…” Ah, well, but philosophers since the Pre-Socratics have understood that empirical evidence (i.e., “science”) can illuminate philosophical questions. However, that’s a far more modest claim than the notion that science can actually answer those questions. Third, Atkins can’t help himself but deliver a contemptuous dig at the “semi-sciences.” This attitude, common among physicists, reflects a naive understanding of the philosophy of science, according to which physics is the (self-professed) “queen” of the sciences, and every other field will achieve full scientific status only when it will finally evolve into something that looks like physics. But an increasingly common view in philosophy is that there actually is a fundamental disunity of science, that “science” is only a loosely defined family resemblance term, reflecting the fact that each science has its own goals, methods, and internal standards, and that there is no universal yardstick to be appealed to in order to make comparative judgments of quality.


Going back to philosophy, the question of “how should I live?” admits of a large number of reasonable (and a lot of unreasonable!) answers, given the very same facts about the universe and human nature. It isn’t so much a question to be answered, as to be explored and clarified. Indeed, this is arguably what most fundamentally distinguishes science from philosophy.
One of my recent morning meditations is pertinent here. It begins with a quote by the Stoic philosopher Epictetus, who says in Discourses II, 11.13:


“Here you have philosophy’s starting point: we find that people cannot agree among themselves, and we go in search of the source of their disagreement.”


As I argue in the podcast episode, there are two broad sources of disagreement among human beings: factual and conceptual. If you and I disagree about, say, the number of moons orbiting around the planet Saturn, one of us is wrong, possibly both. There is a matter of fact about the issue, and we can find out the answer by asking an astronomer. Or more simply by doing a web search. If disagreement remains after that, then one of us is more than a bit obtuse.


The second kind of disagreement concerns how to think about facts, actions, and values. Here the facts are relevant, but insufficient to settle the dispute. Let’s say we have different opinions about the permissibility of assisted suicide. Certain empirical facts are going to be pertinent to the discussion, like information about how the procedure is going to be implemented, what safeguards there may be to avoid abuses, and so forth. But even if we agree on the facts, we may still disagree on the crucial issue: is assisted suicide morally permissible?


That’s the difference between science and philosophy, and why Epictetus says that philosophy begins with the search for why people disagree on things. Notoriously, philosophy does not necessarily settle such disagreements. The joke in philosophy departments is that our profession’s slogan is: “Philosophy: we have all the questions!” But what philosophy does, by means of careful analysis and reasoned argument, is to help us clarify why, exactly, we disagree. That is of huge help to people of good will who wish to honestly pursue discussions in search of better ways to conduct their lives. Atkins may want to take notice.

Biology’s last paradigm shift and the evolution of evolutionary theory – part II

Last time we have seen how evolutionary theory has evolved over the past century and a half, why so many contemporary biologists are calling for what they refer to as the Extended Evolutionary Synthesis (see here and here), and how Darwin, building on David Hume, definitely rebutted the intelligent design argument advanced by William Paley. All as part of a discussion of a paper I published back in 2012, entitled “Biology’s last paradigm shift. The transition from natural theology to Darwinism.” (full text here) In this second part we are going to look at whether the transition between natural theology and Darwinism constituted a paradigm shift, according to criteria laid out by philosopher of science Thomas Kuhn. As I mentioned last time, in the paper I also apply the same analysis to what happened after Darwinism, to more and more recent incarnations of evolutionary theory, but will not discuss that section here.


According to Kuhn, change in science is comprised of two distinct and alternating phases: during “normal science” scientists use the dominant theoretical and methodological tools within a field of inquiry to solve “puzzles”, i.e. problems arising within a particular theory. However, from time to time the number of such problems that cannot be resolved within the adopted framework (“anomalies”) becomes large enough to trigger a crisis, which is then resolved if a new “paradigm” is arrived at to replace the old framework and provide new guidance for further normal-puzzle solving science.


Typically, one of the problems with the Kuhnian approach is that Kuhn did not define exactly what he meant by paradigm, which means that it is not entirely clear what may constitute a paradigm shift. For the purposes of my argument, I will use the commonly accepted interpretation of paradigms as encompassing the “disciplinary matrix,” which means not just the dominant theory or theories within a given field, but also the accompanying methodologies, training strategies for the next generation of scientists, and – no less important – the pertinent metaphysical and epistemological assumptions.


Kuhn suggested five criteria for comparing competing paradigms and for theory choice: 1) Accuracy; 2) Consistency, both internal and with other theories; 3) Scope, in terms of how widely the explanatory reach of a theory extends; 4) Simplicity; and 5) Fruitfulness, in terms of further research. Roughly speaking, then, the comparison between the two paradigms of natural theology and Darwinism is striking. Let’s go through it criterion by criterion.


Accuracy


Natural theology: all explanations are ad hoc, since God’s will is inscrutable.


Darwinism: it can explain some surprising facts about the biological world, like the complexities of the flower structure in some orchid species, or the intricacies of the life cycles of some parasites.


Consistency


Natural theology: internally inconsistent with the idea of an all-powerful, all good God (the problem of natural evil).


Darwinism: as internally consistent as any major scientific theory; external links to other sciences, particularly Darwin’s prediction that the age of the earth had to be greater than what commonly thought by geologists and physicists of the time (turns out, he was right).


Scope


Natural theology: allegedly all-encompassing, but supernatural “explanations” are epistemologically empty. That is, to say “God did it” sounds like an explanation, but it really doesn’t explain anything.


Darwinism: new facts about the biological world that are explained by the theory have been consistently uncovered for more than one and a half centuries.


Simplicity


Natural theology: deceptively simple, if one neglects the obvious question of the origin and makeup of the Creator.


Darwinism: in its original form invokes a small number of mechanisms to explain biological history and complexity; more recent versions invoke more mechanisms, but still a relatively limited number.


Fruitfulness


Natural theology: did not lead to any research program or discovery.


Darwinism: has maintained a vigorous research program for more than one and a half centuries.
According to the above summary, then, the Darwinian paradigm is definitely preferable to Paley’s natural theology – not surprisingly. More interestingly for our purposes here, these are all clear signs of a paradigm shift, the only one ever occurred in evolutionary biology, I argue in the rest of the original paper.


Kuhn’s theory of paradigm shifts famously included another controversial notion: incommensurability, the idea that crucial concepts within a given paradigm are simply not comparable to what superficially appear to be equivalent concepts within another paradigm. Kuhn identified three distinct types of incommensurability: methodological, observational and semantic.


Methodological incommensurability refers to the notion that different paradigms lead scientists to pick different “puzzles” as objects of research, as well as to the idea that scientists then develop distinct approaches to the solution of those puzzles. Obviously, natural theology and Darwinism are methodologically incommensurable: while they both rely on observation and comparative analyses, their goals are entirely different. For Paley, the focus is on the intricate complexity of living organisms, constantly interpreted as an obvious indication of the will and omnipotence of the Creator. Darwin, instead, pays particular attention to precisely those biological phenomena that are troubling to the notion of intelligent design, as in this famous passage:


“I cannot persuade myself that a beneficent and omnipotent God would have designedly created the Ichneumonidæ with the express intention of their feeding within the living bodies of Caterpillars.” (letter collected by Francis Darwin 1887).


More broadly, the sort of “puzzles,” to use Kuhn’s terminology, that Darwinists began to pay attention to concern the historical relationships between different species of organisms (something that is defined out of existence within the natural theological paradigm, since species are specially created), as well as the kind of ecological settings that bring about different adaptations (again, a problem ruled out within natural theology, where adaptations are the direct result of an intelligent act).


Observational incommensurability is tightly linked to the idea that observations are theory-dependent: what is considered a “fact” within one theoretical context may not be such in a different theoretical context. This is perhaps one of the most controversial of Kuhn’s notions, famously illustrated with images from Gestalt psychology, where the same pattern of lines on paper can be interpreted in dramatically different fashions (e.g., a vase or two faces, an old or a young woman, a rabbit or a duck, etc.).


The problem, of course, is that if we take the Gestalt metaphor seriously, we are led to the position that there is no true or even better way to interpret the data, which in turn leads to the constructivist temptation: any theory is just as good as any other, and there really is no way to measure progress in science. Kuhn strongly disavowed such an extreme interpretation of his ideas, and the notion of theory-dependence of observations is now commonly accepted in philosophy of science and embedded in textbook treatments of the subject.


Be that as it may, it is hard to imagine examples of observational incommensurability between natural theology and Darwinism, in part no doubt because no sophisticated way of gathering data was accessible – beyond direct observation and rudimentary experiments – to proponents of the two paradigms.


Finally we get to semantic incommensurability. This has to do with shifts in the meaning of terms used by scientists, one of Kuhn’s examples being the concept of “mass,” which is a conserved, static quantity in Newtonian mechanics, but becomes interchangeable with energy within the framework of Einstein’s relativity.


For the purposes of our discussion, one could make the argument that a similar situation holds for the shifting concept of species between natural theology and Darwinism. Both paradigms do refer to “species,” but the meaning of the term is entirely different. For Paley, species were fixed entities set in place by the action of the Creator – in that sense not far from Newton’s own conception of the physical world, and particularly of the laws governing it. For Darwin, however, species are ever changing entities with no sharp boundaries, which are altered by evolutionary processes in a continuous, gradualistic fashion.


All in all, then, it appears that whether we use the first set of Kuhnian criteria or the various notions of incommensurability, there are very strong reasons to conclude that the shift between natural theology and Darwinism was, in fact, a paradigm shift. It was also, in a very important sense, a shift from a proto-scientific to a scientific view of biology: Darwin and Wallace abandoned any reference to supernatural forces, thus literally establishing a whole new field of science, which keeps, ahem, evolving even today.

Biology’s last paradigm shift and the evolution of evolutionary theory – part I

I find myself devoting more and more time to public outreach and what is increasingly referred to as public philosophy. But, you might know, I am also an academic and publish scholarship in philosophy of science. Which is why from time to time I like to combine the two and do a post that highlights for the general public some of my own technical publications. (A series of these, covering the range 2010-2016, can be found here.) The present essay refers to a paper that I published back in 2012, entitled “Biology’s last paradigm shift. The transition from natural theology to Darwinism.”


The theory of evolution, which provides the conceptual framework for all modern research in organismal biology and informs research in molecular biology, has gone through several stages of expansion and refinement. Darwin and Wallace proposed the original idea back in 1858, centering on the twin concepts of natural selection and common descent. Shortly thereafter, Wallace and August Weismann worked toward the complete elimination of any Lamarckian vestiges from the theory, leaning in particular on Weismann’s concept of the separation of soma and germ lines, and resulting in what is sometimes referred to as “neo-Darwinism.”


The theory then experienced a period of “eclipse” at the turn of the 20th century. Many biologists accepted the idea of common descent but either rejected or greatly de-emphasized the importance of natural selection as an evolutionary mechanism. The situation was exacerbated by the rediscovery of Mendel’s work, which pointed to an apparent incompatibility between discrete inheritance particles (“genes”) and the sort of continuous quantitative variation necessary for natural selection to produce gradual evolutionary change.


Famously, the crisis was overcome with the reconciliation of Mendelian and statistical genetics made possible by Ronald Fisher, J.B.S. Haldane and Sewall Wright in the 1930s, and that later on culminated in the Modern Synthesis (henceforth, MS) of the 1940s, to which several additional authors made important contributions, including but not limited to Theodosius Dobzhansky, Julian Huxley, Ernst Mayr, and George G. Simpson. The MS is still the version of the theory largely accepted by the scientific community, and it is what graduate students in the discipline are trained on.


More recently, several authors have pushed for an Extended Synthesis (henceforth, ES) in evolutionary biology, initially from a variety of individual perspectives largely rooted in particular fields of inquiry, such as Evo-Devo or phenotypic plasticity, and now with a more concerted effort aimed explicitly at the formalization of a broader conceptual framework for evolutionary biology (see here).


The ES is very much a work in progress, but the idea is to accomplish a number of goals that have so far proven somewhat elusive: first and foremost, to finally bring developmental biology – famously left out of the MS – into the fold. Second, to provide a coherent way to reconcile the “holistic” tendencies of organismal biological research with the decidedly more reductionist approach of molecular biology and its most recent products, the various “-omics” (genomics, proteomics, metabolomics, etc.). Third, to incorporate as primary players a number of biological phenomena and processes that had been either discarded or minimized within the context of the MS, e.g., phenotypic plasticity, genetic accommodation, epigenetic inheritance, etc. Fourth, to expand the standard theoretical toolkit of the MS – which is primarily grounded in population and quantitative genetic theory – to include elements from computational biology and complexity theory. Fifthly, to incorporate in evolutionary theory new concepts that have emerged from theoretical research during the past several years, chief among them the triad constituted by evolvability (i.e., the possibility of the evolution of evolutionary mechanisms), modularity (of different components of an organism’s phenotype) and robustness (i.e., the degree of resilience of developmental mechanisms).


In the paper I address the question of whether any of the above amounts to something akin to Thomas Kuhn’s famous paradigm shifts, i.e. whether evolutionary biology has ever undergone anything like what Kuhn describes as a moment of revolutionary science. I argue that it has not, and that it will not, even if the ES will succeed in establishing itself. Rather, I think the only time in the history of biology when such a transition has occurred was during the 19th century, when Darwin’s original theory replaced the dominant “paradigm” of the day, Paley-style natural theology. In the following I will discuss that particular paradigm shift, together with the general Kuhian notion of alternation between revolutionary and “normal” science. People interested in the same analysis applied to what happened to evolutionary theory after the onset of Darwinism are referred to the full paper.


William Paley is responsible for the most articulate defense of the idea that living organisms are the result of a special creation by supernatural forces. In his Natural Theology (1802) he famously introduced the metaphor of a watch and its watchmaker:


“In crossing a heath, suppose I pitched my foot against a stone, and were asked how the stone came to be there: I might possibly answer, that for any thing I know to the contrary, it had lain there for ever: nor would it perhaps be very easy to show the absurdity of this answer. But suppose I had found a watch upon the ground, and it should be inquired how the watch happened to be in that place; I should hardly think of the answer which I had before given, that for any thing I knew, the watch might have always been there. Yet why should not this answer serve for the watch, as well as for the stone? Why is it not as admissible in the second case as in the first? For this reason, and for no other, viz., that when we come to inspect the watch, we perceive (what we could not discover in the stone) that its several parts are framed and put together for a purpose (…) This mechanism being observed (…) the inference, we think, is inevitable, that the watch must have had a maker; that there must have existed, at some time, and at some place of other, an artificer or artificers, who formed it for the purpose which we find it actually to answer; who comprehended its construction, and designed its use.” (1802, p. 5).


Paley argued by analogy from the watch/watchmaker inference to the complex living organism/supernatural intelligent designer inference, despite the fact that such inference had been dealt a devastating philosophical blow by Hume, writing several decades before Paley, in 1779. Contrary to what has been assumed for a long time, Paley appeared to have been familiar with Hume, though he was obviously unconvinced by the latter’s arguments.


Darwin was initially persuaded by Paley’s reasoning, but eventually of course provided the decisive counterpoint that was missing in Hume: an alternative mechanism (natural selection) to generate both biological complexity and the pattern of common descent that was denied by Paley’s natural theology. It behooves us to briefly examine Darwin’s answer, so that we may then proceed to compare Paley’s and Darwin’s “paradigms” in view of Kuhn’s ideas to see whether we can sensibly talk of a paradigm shift occurring at the very onset of evolutionary biology as an independent discipline.


Although Paley is mentioned by name only once in The Origin (on p. 201 of the first edition in the chapter on “Difficulties on Theory”), Darwin mounts a concerted and sustained attack on natural theology in chapters VI and XIII of his magnum opus. Here are some relevant quotes to establish the case. First Darwin explicitly contrasts the type of “explanation” provided by natural theologians with a naturalistic explanation typical of the new science:


“He who believes in separate and innumerable acts of creation will say, that in these cases [of organisms’ behavior that have changed in response to a new environment, without – yet – an accompanying change in the structure of the proper organs] it has pleased the Creator to cause a being of one type to take the place of one of another type; but this seems to me only restating the fact in dignified language. He who believes in the struggle for existence and in the principle of natural selection, will acknowledge that every organic being is constantly endeavouring to increase in numbers; and that if any one being vary ever so little, either in habits or structure, and thus gain an advantage over some other inhabitant of the country, it will seize on the place of that inhabitant, however different it may be from its own place.” (1859, p. 186)


Then he proceeds to directly criticize Paley’s use of analogies to draw a parallel between the inference to human design and the inference to supernatural design:


“It is scarcely possible to avoid comparing the eye to a telescope. We know that this instrument has been perfected by the long-continued efforts of the highest human intellects; and we naturally infer that the eye has been formed by a somewhat analogous process. But may not this inference be presumptuous? Have we any right to assume that the Creator works by intellectual powers like those of man?” (p. 188)


Immediately afterwards, he goes so far as laying out the criteria for the falsification of his hypothesis, in sharp contrast of course with the natural theologian’s ideas, which cannot be falsified:


“If it could be demonstrated that any complex organ existed, which could not possibly have been formed by numerous, successive, slight modifications, my theory would absolutely break down. But I can find out no such case. … Natural selection will never produce in a being anything injurious to itself, for natural selection acts solely by and for the good of each. No organ will be formed, as Paley has remarked, for the purpose of causing pain or for doing an injury to its possessor. If a fair balance be struck between the good and evil caused by each part, each will be found on the whole advantageous.” (pp. 189, 201)


To summarize, then, the idea of intelligent design – which had been around at least since Plato’s Timaeus – had been severely crippled on philosophical grounds by Hume in the 18th century. Still, Paley was able to mount a spirited and detailed defense of it at the onset of the 19th century, while Darwin provided the final blow to it (except of course for the modern resurgence of creationism, which is not an intellectually meaningful movement) for the first time on scientific grounds during the second part of the 19th century. It is on the latter transition that I wish to focus in part II, from the perspective of Kuhn’s paradigm shifts.


(next: the Paley-Darwin transition as a paradigm shift)

Is exposing medical pseudoscience unethical?

You would think the answer to the title question is obvious: no, of course it isn’t unethical to expose medical pseudoscience (or any kind of pseudoscience). How can anyone think so? And yet, there are some decent reasons to doubt the certainty of such a conclusion. At least that’s the take of a paper that so far as I know has not yet been published, but is available at the PhilSci (philosophy of science) Archive at the University of Pittsburgh.


The paper is entitled “Exposing medical pseudoscience may be unethical,” and is authored by Ehud Lamm. Here is Lamm’s argument, as he presents it:


(1) Many ill health situations, in particular those involving multi- and comorbidity do not have very effective treatment options.


(2) Alternative medicine, consisting of various pseudo-scientific enterprises, provides repose to some of those suffering from such conditions.


(3) A significant part (if not all) of this benefit can be attributed to the placebo effect.


(4) In current, “Western,” societies, placebos are most effective when they have the symbolic currency of science.


(5) Ergo, exposing these pseudo-sciences will hurt people for whom there are only limited health care alternatives.


This is a reasonably constructed argument, so if one wishes to reject it — as I do — one needs to attack one or more of the premises, or to uncover and then reject some additional, hidden premises. I will follow both strategies here, though I wish to thank Lamm at the onset for having forced me to think more carefully about something I had hitherto taken for granted.


I am going to agree right away with premise (3): most (indeed, very likely, all) of the benefits of alternative medical treatments are the result of placebo effects. But I have varying degrees of issues with premise (1), (2), and (4). And I also think Lamm is not explicitly listing a few other very relevant premises here, though he mentions one of them in passing in the paper, as we shall see, only to discard it far too quickly.


Premise (1): this threads on an ambiguity surrounding the word “treatment.” Yes, modern medicine does not provide effective treatment against certain conditions, for instance a number of types of terminal cancer, or several types of dementia. But alternative medicine does not provide treatments either, according to Lamm’s own premise (3), only palliatives. Which means that the proper comparison here ought to be only between standard palliative care and “alternative” one. Palliative care works on the basis of a combination of actual biomedical effects (e.g., painkillers) and placebo. Other things being equal, then, it is more effective for a doctor to use her prestige (premise 4) to push tested palliative options rather than pseudoscientific ones.


Premise (2): for the reasons just seen, even if alternative medicine does provide some respite to patients by way of a placebo effect, standard palliative care is in any case preferable, because it also works by placebo, but additionally actually delivers (at least in some cases) a real biomedical effect. Again, remember that we are talking about alleviating pain and similar things, not actual cures, which are not achievable via placebo, and Lamm agrees that placebo is pretty much the only mechanism through which pseudo-medicine “works.”


Premise (4): the symbolic currency of science in the Western world seems much less real than Lamm assumes. Witness the widespread antivax movement, or — outside of medicine — so-called “skepticism” of climate change, or of evolutionary theory. If anything, science is suffering a credibility crisis of late, and inducing medical doctors to deploy pseudoscience is hardly going to improve such credibility.


If the reader agrees with even one of my criticisms of Lamm’s premises, that is enough to reject the argument. But there is more.


For instance, Lamm at one point says: “I am putting to one side deontological and consequentialist arguments against the use of placebos in general, and assume that in some situations relying on a placebo effect is legitimate.” That’s a bit too quick, and a rather big thing to put aside (and the bit after the comma somewhat begs the question)! A deontologist, for instance, might argue that it is a violation of the categorical imperative for a doctor to systematically lie to his patients, because that violates the implicit trust between the two (the very same trust on which premise 4 is built, incidentally), and because the doctors themselves would probably rather not been lied to when it is their turn to be patients.


On consequentialist grounds, one could argue that there are long term negative societal consequences engendered by lying to patients and by pushing pseudoscientific notions. Because, again, patients might eventually lose confidence in their doctors and begin to doubt any sort of medical advice; also because people will be more likely to embrace pseudoscience under other circumstances, for instance when there are, in fact, perfectly good evidence based treatment options available; and mroevoer because we would be condoning a multibillion dollar industry based on what is essentially a fraud.


Furthermore, there is a third big ethical framework out there: virtue ethics. It is hardly a good character trait to engage in systematic deception of others, even for their own good (a premise, this last one, that I have already rejected). Virtue ethics does not follow universal rules, so lying can be acceptable under certain circumstances. But the focus is on the character of the moral agent, and repeated lying is going to be deleterious to that character, since character is shaped precisely by repeating the same actions over and over, or by systematically entertaining and agreeing with the same thoughts.


Lamm sees another concern with exposing pseudo-medicine: “the movement toward Evidence Based Medicine (EBM), to the extent that it is successful, may lead people to be less open to treatments and assistance that are not backed by science, such as talk therapies, meeting with alternative medicine practitioners, support from informal social networks, help from clergy, and so on.”


This is an interesting point, but the way it is articulated it carelessly lumps a number of things that should be considered separately. For instance, homeopathy is clearly and definitely a pseudoscience, so it should be counted as a success if people do not waste their money (and hopes) on very expensive sugar pills. Talk therapies, however, are not necessarily pseudoscientific. We have good evidence that some forms of psychotherapy work (e.g., cognitive behavioral therapy), and there are distinctions to be drawn about what “working” means here. If a patient is in need to simply talk to someone, but is under no illusion that this will actually cure him, I don’t see what the problem is, or why such talk therapy should be counted as pseudo-medicine in the first place. Perhaps it won’t work better than getting a drink with your friend, but if chatting is the only thing one needs in order to feel better, it “works.” Much hinges, of course, on what the precise claim of the therapist is going to be. The very same considerations apply to the use of informal social networks, or to help from the clergy. These things are simply not in the same category of homeopathy and other kinds of pseudo-medicine.


And there is more to consider, like the general ideas that truth and honesty are intrinsically valuable, and should be compromised only under very unusual circumstances. Certainly both a deontologist and a virtue ethicist would reason this way, though a consequentialist might disagree (there goes yet another reason to reject consequentialism, in my view).


Lamm does consider some objections to his argument in the second part of the paper, and rejects them with a number of counterarguments. I have already covered part of what he says there in the previous points, but let me add a few pertinent comments.


He writes: “we should strive for alternative channels of social support to be available and determine whether non-science backed alternatives can provide patients with social and economic benefits that they need, such as health insurance coverage, help with the education system, and so on.” Well yes, as acknowledged before, but again this thread on an ambiguity of the term “support.” There is no problem in supporting people via talk therapy, social networks, religious counseling, and so forth. But these are not types of alternative medicine, and so long as the people providing the support are not lying, then there is no ethical issue. (Of course if you are an atheist you will think that all priests are lying by definition, but this is not the case from the point of view of a believer who has already adopted that particular religion’s framework. Admittedly though, that’s a tough case, best explored on its own.)


Lamm also maintains that some pseudoscientific “treatments” may be cheap and effective (again, via placebo) and therefore preferable whenever there are no established medical options for the condition. But we have seen above that at a very minimum modern medicine can provide palliatives, and let’s remember that much alternative medical practices are anything but cheap.


The author acknowledges a big potential downside of his approach: “if the pseudoscience providing the placebo effect causes patients to ignore science backed treatment we have to consider the tradeoff to decide whether belief in the efficacy of pseudoscience is beneficial or harmful.” Indeed, and this may often be the case. Not to mention that so long as the patient is also seeking regular medical treatment then, again, there is no additional benefit of the pseudo-medical one, which will cost money not just to the patient, but also, increasingly (and unfortunately) to the health insurance or government-backed apparatus that will pay for the so-called treatment.


Finally, Lamm argues: “It may be mistakenly suggested that the argument presented here is a justification for fraud (assuming the pseudoscience in question involves fraud). However, if anything, the argument discussed here only applies to exposing fraud, which in some cases may do more harm than good.” I don’t see how this would work: if a medical practitioner knows that a given remedy is pseudoscientific and recommends it nonetheless, he is participating in a fraud, even in cases in which he is not directly pocketing money for it (someone else is, and besides, he is being paid to take care of the patient). So, refraining from exposing fraud is, seems to me, condoning and abetting fraud no matter how one puts it. This may be a case of philosophical distinction without a difference.


Again, I recognize that Lamm’s paper is important because too often people like myself, who have pursued a career of exposing pseudoscience and defending science, make leaps of reasoning from the factual to the ethical, automatically assuming certain value judgments without further argument. But it seems to me that, once the arguments have been examined, there is still a strong preponderance of reasons to expose pseudo-medicine whenever possible.