Evolutionary Theory as a Theory of Forces

In a post a few weeks back, I expressed skepticism towards labeling evolutionary processes such as selection, drift, migration and mutation as “forces.” At this point I’m not even sure what to call them – “processes” doesn’t seem right either. So I decided to read up on the topic.

As far as I know, the first formulation of selection, drift, migration and mutation as forces is Elliott Sober’s “Evolutionary Theory as a Theory of Forces,” the first chapter of his book, The Nature of Selection (1984). The chapter serves both as an argument for treating evolution as a theory of forces as well as introducing some of the basics of evolutionary theory. I will try to distill the “theory of forces” in this post.

While Sober never explicitly says what exactly a “theory of forces” entails, he mentions multiple prerequisites throughout the text. These include vector quantities that can resolve in a single direction, a zero-force state, and a differentiation between source laws and consequence laws. I will first describe these requirements specifically and show how they are reflected in a Newtonian theory of forces. I will then describe why Sober believes evolution can be seen as a “theory of forces.”

Vectors are quantities that have both a magnitude and a direction. This is the difference between speed and velocity – speed is a magnitude, velocity is speed with a direction. Furthermore, vectors can be added and subtracted which results in the resolution of component forces – multiple forces result in a singular net force. If I push an object north, and you push an object east, the pushes (component forces) are added together (resolved) and the object travels northeast (a net force).

The zero-force state describes the object when either no forces are acting on the object or the acting forces cancel each other out (like two pushes of equal force, one east and the other west). This is Newton’s First Law (or the Law of Inertia) in a nutshell – an object travels with constant velocity unless acted upon by a net force. In a way, the zero-force state can be viewed as the null hypothesis.

Another requirement is source laws and consequence laws. I’ll quote Sober here:

A theory of forces must contain both source laws and consequence laws. The former describe the circumstances that produce forces; the latter describe how forces, once they exist, produce changes in the systems they impinge upon. The classical law of gravitation is a source law; it says that when two objects of given mass are separated by a given distance, there will be a gravitational force of certain magnitude (50). (italics original)

He subsequently says that a consequence law is exemplified by F=ma. “No mention is made here of the physical conditions that produce forces, but only of what forces do, once they exist” (50). F=ma doesn’t tell us anything about the nature of the force, like a gravitational force, but F=ma can be used to compute the consequence of the gravitational force.

How does Sober think these requirements are reflected in evolutionary biology?

Well, when evolution is defined as “change in allele frequencies,” selection, drift, migration and mutation can be seen as “pushing” allele frequencies. These pushes have a magnitude and a direction (either the frequency increases, decreases, or stays the same). I feel a little weird about this, however, as these “pushes” are theoretical pushes. Perhaps when an evolutionary landscape is considered, this view makes more sense. I don’t know much about the landscape though so I can’t comment any further.

What is evolution’s zero-force state? Sober says it’s Hardy-Weinberg equilibrium (34). Generally, Hardy-Weinberg describes what will happen to allele frequencies in the absence of a net force, in that when there are no “forces” acting upon allele frequencies, allele frequencies will reach an equilibrium and subsequently no longer change (and as far as I understand, equilibrium is reached within one generation of zero “net force”). This seems like a reasonable zero-force state.

What about source laws? Sober claims that biology has a nearly unlimited amount of source laws since the physical conditions that cause zebras to evolve are different from what cause goldfish to evolve. However, biologists can discover source laws that are slightly more general than a case-by-case basis – Sober argues that sex ratio is one of these (51). In a population with an uneven sex ratio, parents who produce offspring of the minority sex will have a higher fitness than those that produce offspring of the majority sex. Indeed, this is generalized scenario that can apply across taxa. Unequal sex ratio can be seen as a source law – the “circumstances that produce forces,” in this case, natural selection, the consequence law (59).

Interestingly, Sober sums up source laws and consequence laws this way: “Whereas it is mainly ecology that tries to produce source laws for natural selection, the consequence laws concerning natural selection are preeminently part of the province of population genetics” (59). In other words, ecology discovers why there are variations in fitness, and population genetics takes those differences and puts them through equations (but ultimately doesn’t care what caused the variations in fitness), akin to what brings about a gravitational force and F=ma.

This seems convincing. I am certainly more convinced after reading the actual arguments rather than just saying “No, this doesn’t make sense!” after reading a sentence or two about it. I still reserve some skepticism, however, and will continue to read the literature on this topic (it seems Sober’s argument has provoked quite some debate!) and update the blog as I progress. If anyone has any input or good sources to read on the topic, I would appreciate the comments!