One of my readers, Jake, recently asked whether I was doing good science by relying on metaphor (see comments in previous blog post). Initially, I thought that I would reply by defending the use of metaphor. Upon reflection, however, I realized that I was not using metaphor more than those who I was arguing against. So, my reply has changed from a defense of metaphor to a defense of the metaphors that I employ.
Modern evolutionary theory relies on metaphor as follows. It defines natural selection as heritable variation in fitness. This definition of natural selection redefines fitness from adaptedness to net reproductive success. Relevant to metaphor, this definition makes an implicit comparison between something that is difficult to model mathematically, and which varies from species to species— a complex organism’s multidimensional adaptedness to its environment—and something that takes a very simple mathematical value across taxa—the net number of offspring an organism produces. Fitness is thus used in a way that differs from its normal English usage to suggest a resemblance between two things, adaptedness and reproductive success. An organism can be highly adapted to its environment, and still fail to reproduce (or die). In this sense, the modern fitness concept is a metaphor and undergirds the modern definition of natural selection.
By focusing on measures of reproductive success, the modern fitness-based theory can be useful for analyzing the relative importance of mutation, drift, and selection, and other “genetic” factors independent of the particular historical context. On the other hand, it cannot say much about the origin of particular complex traits. For that, we need Darwin’s conceptualization of natural selection, which disentangles measures of net reproductive success from organismal adaptedness. Under Darwin’s approach, a much greater focus is placed on fit between organism and environment, both in contemporary populations and extinct populations. How many component parts does a trait have? How might those parts have each evolved separately, and come together during history? Are there fossils or comparative data that give clues to particular transitional states?
Many of the sorts of questions that Darwin addressed can be answered without knowing anything about how a trait might have affected survival and reproduction in some previous context. When investigating adaptation, a first step is breaking a complex trait into simpler components. A second step involves figuring out how sub-components interact and the sequence of steps through which they may have been assembled. A third step is figuring out at what point a complex trait yielded a breakthrough function. A final step may include investigating how particular sub-traits affected survival and reproduction in ancestral environments, and whether the minute selective factors leading to a breakthrough trait are the same or different from its impact on success. In other words, we can learn a lot about adaptation, from Darwin’s approach, before we ever get to questions of survival and reproduction. In contrast to modern workers, Darwin did not redefine fitness in a way that confuses adaptedness and reproductive success, because he often independently studied adaptedness.
In my paper (Gilbert 2020), I employed Darwin’s conceptual scheme, and I argued that Darwin’s metaphors can be extended to incorporate macroevolutionary processes. To summarize some of my various arguments:
- Darwin defined the levels of selection by reference to the struggle for existence metaphor. Under Darwin’s argument, most competition relevant to natural selection happens within species, not between species or higher taxa. Moreover, because competition is between forms that differ subtly from each other, changes are gradual and happen in small steps.
- Building actual models for the origin of complex traits shows that they often originate randomly with respect to long-term success (Gilbert 2015, 2017, 2020).
- With random variation at higher levels of the genetic hierarchy, there can also be a different form of competition in evolution, which happens as a race to innovate. Here, the first forms to exploit an untapped resources are naturally rewarded with an incumbent advantage (Gilbert 2020).
- Natural selection happens under effective resource limitation, and the struggle for existence involves changes of types within a species.
- Natural reward happens under effective resource abundance, and the struggle for supremacy involves periods of transient population increase, which alters the total abundance of genetic systems as shared by higher taxa.
- The metaphor of “natural reward” suggests a resemblance between the way that untapped markets in human societies reward innovation, and the way that untapped resources in nature reward innovation (Gilbert 2020).
- Under a theory that includes natural reward, innovation can be defined as including invention (the origin of complex traits) and entrepreneurship (their spread); progress can be defined in terms of population expansion, increased biomass or energy flux, depending on what is being compared; and advancement can be defined as the increase of innovativeness (Gilbert 2020).
- Natural reward is expected to lead to the success of the innovative, progress, and advancement over time.
Jake also cited the hydraulic metaphor of electrical energy flow to expose the limitations of metaphor in science. However, just as the hydraulic metaphor fails to capture that electrical energy flows as a field surrounding the wires and not within the wires, modeling natural selection as heritable variation in fitness fails to capture that relevant process is within species and not above the species level. This is also true if we define natural selection as differential survival and reproduction of individuals due to differences in phenotype, as Jake did, without specifying the levels of differential survival and reproduction (i.e. within or between species?). If we understand natural selection as a within species process, then we can derive useful mathematical theories—in the same way that we can derive useful mathematical theories of electrical energy flow if we use field models. However, we need Darwin’s struggle for existence metaphor to understand why natural selection applies within species only. The modern definition of natural selection as heritable variation in fitness leaves the door open to selection on any level with apparent reproduction, including superspecific levels.
At present, we have more formal models of the origin of complex traits, than we have models that incorporate the effects of random variation of invention, and differential transient population increases, on major macroevolutionary patterns. But we can use new metaphors as to guide new investigations. Gooding (1980) argued that, “Analogical reasoning enabled Faraday to translate deeper assumptions about the unity or correlation of forces into experimentally realizable forms without resorting to hypothetical models about underlying structures.” Likewise, analogical reasoning enabled Darwin to study microevolution without knowledge of inheritance mechanisms. We may forge a foundation for a new theory even before we derive formal mathematical models.
In my work, I returned evolutionary theory to core Darwinian foundations, and then extended it in a new direction. My goals is to arrive at accurate knowledge, and my theory allows useful definitions of words that previously remained ill-defined. My hope is that this theoretical foundation will allow for the construction of mathematical models of long-term processes and an exploration of the utility of definitions for words like progress and advancement. I am concerned with answering questions about evolution at the broadest scales. In seeking this goal, I do not aim at precision and mathematical tractability at the cost of explanatory power. My goal is to seek the truth using whatever methods are appropriate. At the early stages of developing a new theory, I feel that this requires a grasp on the concepts first.
Acknowledgments: I thank Jake PA for stimulating discussion.
References
Gilbert, O. M. (2015). Histocompatibility as adaptive response to discriminatory within-organism conflict: A historical model. The American Naturalist, 185(2), 228-242.
––– (2017). Association theory: a new framework for analyzing social evolution. bioRxiv, 197632.
––– (2020). Natural reward drives the advancement of life. Rethinking Ecology, 5, 1.
Gooding, D. (1980). Faraday, Thomson, and the concept of the magnetic field. The British Journal for the history of Science, 13(2), 91-12