Hull’s Science as a Process (U. Chicago, 1988) invited us to borrow a biological theory, natural selection, after philosophical streamlining, to analyze scientific activity and conceptual change. This ambitious synthetic work of the late David Hull, a philosopher of evolution and biological systematics, had been the subject of many reviews and symposia since it appeared (see e.g. Angier, Ruse, Maynard-Smith, Kirsch, Latour) at the time I prepared this commentary (1993). Many commentators had been hard on Hull’s sociology and his credit economy. Less attention was paid to the conceptual foundation of this work, namely, the borrowing of a biological theory, natural selection, and its deployment, after philosophical streamlining, to analyze scientific activity and conceptual change. Continue reading
How does this happen? How often does this happen? might be a response to the earlier post citing Waddington’s experiments, in which variation that originated as an appropriate response to environmental circumstances became more or less fixed over time in a population. To answer the “how often” question, researchers have to be looking for examples, and to look for examples they not only have to be motivated but to have a model of what they are looking for, that is, an idea of how it happens. First, how it happens in my view, then in the view of Waddington and others.
Consider some acquired character, moreover one that is an appropriate response to the environmental stress. Likely there will be a range in responsiveness. Moreover, the developmental paths to the “same” response will typically differ among individuals. If there is any survival and reproductive advantage to greater responsiveness, the offspring of matings in the population will come to concentrate in a subsection of the space of developmental paths. We can expect that eventually some individuals in that subspace of responsive developmental paths will show the character in the absence of the environmental stress. That is not always a good thing, but, if it is advantageous, the subspace of responsive developmental paths will come to include ever more of those “pre-responsive” individuals. The critical part of this process—some offspring in the “concentrated” subspace of responsive developmental paths showing the character in the absence of the environmental stress—is not as a logical matter, but as a consequence of development. No reverse transcription in individual transmission is needed in this Darwinian, population-thinking version of Lamarck.
If this is how this happens, I concede that we do not have much evidence to infer that this phenomenon is widespread in nature. However, until more researchers can conceive how it can happen, they won´t be looking for it, and so evidence for it will remain slim even if it occurs often.
There are other accounts of how this happens. In Waddington´s own terms the previous environmental induction of the characters has been taken over by a genetic switch and become “genetically assimilated” (Waddington 1961). More conventional neo-Darwinists than Waddington have obscured the implications of his experiments by inventing, without further investigation, gene-centered explanations. GC Williams, in his influential book, Adaptation and Natural Selection, attributed the fixation of acquired characters in Waddington´s experiments to a change in the balance between a fixed number of suppressor alleles and a variable number of positive alleles accumulated under natural selection. (Similarly, see Mayr 1970, 361-5.) The attention is diverted away from the non-random origin of the responses and onto the postulated suppressor or positive alleles, which presumably arise randomly (that is, without reference to the environmental stress) and are then naturally selected.
The random origin of characters and the conceptual separation of origin from their change in frequency in a population makes it difficult for evolutionary theory to give significance to the structured activity of organisms during their lifetime, something that may influence the direction of change in a population. Addressing that blindspot is important whatever the answer turn out to be to the question of how often variation that originated as an appropriate response to environmental circumstances becomes more or less fixed over time in a population.
This post is adapted from P. J. Taylor, “Historical versus Selectionist Explanations in Evolutionary Theory” Cladistics 3: 1-13,1987.
Mayr, E. 1970. Population, Species, Evolution. Cambridge: Harvard University Press.
Waddington, C. H., 1961. Genetic assimilation. Advances in Genetics, 10: 257-290.
Williams, G. C. 1966. Adaptation and Natural Selection. Princeton: Princeton University Press.
When Darwin, in the third chapter of On the Origin of Species, explored evolution’s ecological context he was not simply laying out a program of research for a future science now called ecology (see previous post). He was responding to anticipated criticisms of his theory of natural selection as the mechanism for evolutionary change that produces the “that perfection of structure and co-adaptation which most justly excites our admiration.” Continue reading
Integrating the structure and dynamics of evolution’s ecological context (see previous posts) remains a neglected project within evolutionary theory. Nevertheless, the different approaches to theorizing ecological organization can still be read in terms of the ways that evolutionary theory fits into them, whether or not this is made explicit. Table 1 provides a classification of five basic orientations.
Central to the first three orientations is the notion of system, which I use in the strong sense of an entity that has clearly defined boundaries and has coherent internal dynamics, dynamics that govern the system’s responses to external influences and determine its structure, stability and development over time (Taylor 1992). System in this sense can refer not only to the basic units of systems ecology, but also to the guilds and communities of community ecology. These three orientations differ according to the relative time scales of ecological and evolutionary processes. In contrast to viewing ecological organization as system-like, various ecologists have emphasized what I call its “unruly complexity” (Taylor 2005). That is, organisms and processes transgress the boundaries of any unit of ecological structure, spanning levels and scales; natural categories for and reduction of the complexity are elusive; ecological structures are subject to restructuring; control and generalization are difficult. The two non-system orientations differ according to whether this unruly complexity can be disciplined theoretically. Table 1’s distinctions are illustrated in Taylor (2000) through a review of twentieth century theories of ecological organization.
In the next post in the series, I note Darwin’s keen awareness of the structure and dynamics of evolution’s ecological context and mention some research that follows in that tradition.
Table 1. Five orientations to theorizing ecological organization and evolution.
|system (or community)||system evolves as a Coherent whole||Fast return to equilibrium; slow change or evolution of system|
|individuals in context of system||Stable system||Fast return to equilibrium
intermediate speed evolution of population of individuals
slow change of system
|system transient, yet Regularly reoccurring||Fast passing of transient context (e.g.,succession)
intermediate speed evolution of population of individuals
slow change in nature of transient context
|ecological organization as not system-like||Anti-Theory||Relevant processes not separable into “ecological” and “evolutionary” time scales|
|unruly complexity can be Disciplined|
Taylor, P. J. “Community” pp. 52-60 in E.F. Keller & E. Lloyd (eds.) Keywords in Evolutionary Biology, Harvard University Press, 1992
—- “From natural selection to natural construction to disciplining unruly complexity: The challenge of integrating ecology into evolutionary theory,” in R. Singh, K. Krimbas, D. Paul & J. Beatty (eds.), Thinking About Evolution: Historical, Philosophical and Political Perspectives, Cambridge: Cambridge University Press, 377-393, 2000.
—- Unruly Complexity: Ecology, Interpretation, Engagement Chicago: University of Chicago Press, 2005.
An example of the contribution of developmental and ecological flexibility to the evolutionary origin of characters (see previous post) involves barn owls recently migrated to Malaysian oil palm plantations. Lenton (1983) describes the owls fifteen years after the first pair bred in a Malaysian plantation, by which time they had spread throughout southern Malaysia. The owls have two to three clutches a year, do not rigidly defend their territories, and perch to wait for their prey-the rats abundant in the plantations. In contrast, owls of the same species in Europe have only one clutch per year, rigidly defend their territories, frequent open areas, and quarter those areas in search of prey. Moreover, the juveniles in the Malaysian plantations, before the age of establishing their own nesting sites, congregate at the end of the day’s hunting-social behavior not observed elsewhere.
Now, if Darwinian biologists first observed the character differences between the Malaysian owls and their relatives without knowing about the recent immigration, my guess is that they might explain the increased fecundity and contracted territories as selectively advantageous in the environment of abundant rats. They would assume that there must have been some variation in fecundity and territory size in the original owls. The juvenile congregation would probably be passed over as simply a by-product of the other selected changes.
Suppose we then informed the biologists of how few generations there had been to accumulate any differentials that could have emerged from the offspring of the one founding pair. They would probably shift the focus of their explanation and postulate that barn owls have the flexibility to develop responses to novel environmental circumstances in appropriate ways, even if we have not observed the responses elsewhere. Flexibility would be seen as an adaptation resulting from previous selection. But I doubt that modern Darwinian biologists would go further and conclude that the owls have in fact evolved and adapted without any significant genetic change in the population.
Even if the biologists did not modify their definition of evolution to incorporate non-genetic change, the case indicates how the historical conjunction of circumstances and the previously unobserved ecological flexibility of the pre-immigrant owls elicited the new characters. This conjunction becomes crucial to the explanation of the changes that were observed in the population of oil palm owls—as important as, perhaps more important than, the genetics of character variation or any differential representation among variants after migration into Malaysia. Understanding evolutionary change does not license our focusing on the characters of individuals and not attending to the dynamic relationship between individual and contextual change.
Development is central to this story. Instead of single characters directly linked to genes, typically hypothetical, that arise through mutations or random rearrangement of DNA, we see that new characters arise within integrated sets of characters that develop over the organism’s lifetime. Instead of inheritance as the transmission from parent to offspring of Master Molecules, we have a picture of characters always being reproduced (imperfectly) through flexible developmental processes, processes that are only conditioned, not determined, by genes.
In the next post in the series I return to the original question of what it might look like to make evolutionary studies more ecological.
Lenton, G. (1983) “Wise owls flourish among the oil palms,” New Scientist, 97: 436-437
An extract from Taylor, P. (1998) “Natural Selection: A heavy hand in biological and social thought,” Science as Culture, 7 (1), 5-32.
The last post identified special conditions that increase the chances of natural selection (carefully construed) serving as an explanation of the historical change in the frequency of one character. From a knowledge of biology, we should agree that these special conditions are rare or not necessarily generalizable. The consequences for evolutionary theorizing have been several. Biologists (and others) often:
- collapse “selection,” using the term as a synonym for differential representation of characters[i]
- rely on claims about current functionality without evidence of historical (temporal) change;
- accept milder standards of evidence (e.g., the historical process has been observed in some cases of natural selection, so it is plausible that it occurred for the character whose current function has been demonstrated; or one works at a coarse level of resolution of characters, environments, and change so that departures of the detailed mechanisms from the special conditions are not evident[ii]);
- invoke repeatedly the same few textbook cases of natural selection; or
- perform laboratory or other experimental work in which selection literally, not metaphorically, takes place.
I concede that applying the strong standards of evidence I have outlined may result in few natural selective accounts qualifying as adequate historical explanations. Nevertheless, it should also be recognized that squeezing evolutionary change so it can fit the special conditions above has the effect of discounting many important aspects of biology:
- characters that are not singled out in living activity of organisms;
- the development or the reconstruction during an organism’s lifetime of its characters, over and above the transmission of genetic and other material to the zygote (and in contrast to snapshots of characters at some point of time in the lifecycle);
- the broader conditions for “recurrency” of characters, which depend not only on the genetics implicated in the development of characters, but also on the persistence of environmental conditions at least insofar as the organisms modulate or “construct” those conditions (Lewontin 1982, 1983, 1985);
- the contribution of developmental and ecological flexibility to the evolutionary origin of characters (Taylor 1987)[iii]; and, more generally,
- the structured, yet changing, ecological dynamics to which organisms both respond and contribute.
In short, characters are part of structured processes. For some biologists and philosophers of biology these complexities of biology mitigate against the coherent accumulation of change over time; they believe that only when the special conditions more or less apply does evolution lead to identifiable adaptive outcomes. Others attempt to incorporate some of these aspects of biology by adjusting the theory of natural selection, as in, for example, frequency-dependent selection. This tinkering, however, preserves room for the almost conventional moves back and forwards in evolutionary thought among forward speculation and backward fitting.[iv] My preference is to free ourselves from the restrictive, and thus widely misused, form of the natural selective explanation.[v]
In the next post in the series I include an example of the contribution of developmental and ecological flexibility to the evolutionary origin of characters. Then I return to the original question of what it might look like to make evolutionary studies more ecological.
Keller, E. F. and E. A. Lloyd (Eds.). (1992). Keywords in evolutionary biology. Cambridge, MA: Harvard University Press.
Lewontin, R. C. (1982). Organism and environment. In Learning, Development, and Culture, ed. H. C. Plotkin, pp. 151-170. New York: John Wiley & Sons.
Lewontin, R. C. (1983). The organism as the subject and object of evolution. Scientia 118: 63-82. Reprinted in The Dialectical Biologist, ed. R. Levins and R. C. Lewontin, pp. 85-106. Cambridge, MA: Harvard University Press.
Lewontin, R. C. (1985). Adaptation. In The Dialectical Biologist, ed. R. Levins and R. C. Lewontin, pp. 65-84. Cambridge, MA: Harvard University Press.
Taylor, P. J. (1987). Historical versus selectionist explanations in evolutionary biology. Cladistics 3: 1-13.
Another extract from “From natural selection to natural construction to disciplining unruly complexity: The challenge of integrating ecology into evolutionary theory,” in R. Singh, K. Krimbas, D. Paul & J. Beatty (eds.), Thinking About Evolution: Historical, Philosophical and Political Perspectives, Cambridge: Cambridge University Press, 377-393, 2000.
[i] A variant of this is the idea that a character could be the cause of natural selection for organisms having that character if the character has a positive effect on survival and reproduction averaged over the range of contexts in which the character occurs. This conceptual move underwrites theories about natural selection of sub-organismic units and sometimes, super-organismic units.
[ii] In the hypothetical example in an earlier post in the series, if we had not noticed the plant’s relative that had the same angle without the hummingbird pollinator, but we were aware of the ancestor, natural selection would have been both functionally and temporally plausible. At some point, however, we balk at allowing a lessening of resolution to support natural selective explanation. We know, for example, that it is too coarse to correlate bird feathers both functionally and temporally to the bird lineage’s move into the air.
[iii] In order to fully account for the direction of the observed historical change one should—even in a natural selective explanation—study the origin of the character. This explanatory requirement might account for biologists’ reliance on another special condition, namely, the origin of characters by mutation or recombination, which make this origin random with respect to environmental circumstance.
[iv] As evident, in particular, in debates about “fitness,” units of selection, and levels of selection (see relevant entries in Keller and Lloyd 1992).
[v] It follows that I also propose abandoning the concept of adaptation, in both its senses, i.e., the character that has been the causal focus of a natural selection account and the process of its evolutionary development. My thinking along these lines drew at a key stage on Lewontin’s critique of the concept (see Lewontin 1983).
What is needed to demonstrate that some evolutionary change and the resulting characters were produced by a process of natural selection? The short answer (from the previous post): It’s hard work to establish evidence for natural selection. Longer answer: It requires demonstrating both a functional and a temporal correlation between the character and the differential reproductive success, i.e. analysis of the character’s effect and its consistent origin in time with respect to the environmental circumstances/ stress/ challenge.
In practice, most evolutionary biologists or writers about evolution don’t undertake that work. They seek out or imagine special conditions that increase the chances of natural selection serving as an explanation of the historical change in the frequency of one character:
- a) Single function (or co-ordinated composite function) for the character (so you can discount other effects in the functional correlation).
- b) Effective independence of the character and its effects from those of others (again so you can discount other effects in the functional correlation); e.g heavy metal tolerance in plants on mine tailings.
- c) Character reproduced over time (generations) (so you don’t have to deal with changing characters and their functional correlations); e.g. single locus phenotypes.
- d) Consistency of the organism-environment relationship over time (again so you don’t have to deal with changing functional correlations).
- e) Limited time span (again so you don’t have to deal with changing functional correlations).
From our knowledge of biology, these special conditions are likely to be rare. The next post addresses the consequence of their rarity.