(This is a summary of what Alex Rosenberg and Daniel McShea say about biological laws in their book “Philosophy of Biology”. I think this book is intended to be an introductory book of philosophy of biology and I don’t think what they say here necessarily expresses both of their views, e.g., I think Rosenberg does hold that natural selection is a law of nature.)
It’s hard to see how there are any laws in biology in the way there are chemical and physical laws. Some may think that this is a defect that needs to be repaired. But most think that the lack of laws is because biology is more difficult and different than physics. In biology there are mathematical models, e.g., Mendel’s laws, Fisher’s sex ratio model, and Hardy-Weinberg equilibrium. The question is: are these laws similar in kind to the laws in physics?
In physics, a theory is a body of scientific laws that work together to explain phenomena. Theories require sets of laws because explanations identify causes, and causes are a matter of lawful regularity.
Laws must have the form “If P then Q”, or equivalently, “All Fs are Gs”, where F and G are events, things, processes, properties, etc. Law must also be universally true, everywhere and always.
Rosenberg and McShea offer four criteria for laws.
- Is the candidate a true universal conditional that makes no mention of specific places, times or things?
- Is the candidate a contingent statement the denial of which is conceivable, as opposed to a definition or the consequence of definitions that cannot report causal relations?
- If the candidate is true only because of a ceteris paribus statement, can we expect to narrow the range of its exceptions by empirical means?
- Does the candidate support counter-factual conditional statements?
Let’s consider Mendel’s laws, the Hardy-Weinberg law, and Fisher’s sex ratio model.
Mendel’s law of segregation: In a parent, the two alleles for each character separate in the production of gametes, so that only one is transmitted to each individual in the next generation.
Mendel’s law of independent assortment: The genes for each character are transmitted independently to the next generation, so that the appearance of one character in an offspring will not affect the appearance of another character.
Mendel was lucky that the traits he first studied did not have genes that were linked close together on the same chromosome. Otherwise, the genes would not have assorted independently. Biologists today know that Mendel’s second law is a generalization that has an enormous number of exceptions. There are also exceptions to the law of segregation, where one of the two alleles is preferentially transmitted to the next generation (segregation distorter alleles).
Whether Mendel’s principles are successfully applied in prediction depend on more fundamental regularities in meiosis and other details in cell physiology. But there are no fundamental laws of biology that explain Mendel’s generalizations, because we could imagine that natural selection could operate in other worlds with replicators and interactors that do not use meiosis or other Earthly sexual physiological processes. Any processes in biology could be limited to a certain place on Earth, and at a certain time on Earth.
In light of considerations like this, some philosophers of biology argue that biological laws are different than laws in chemistry and physics. If biological laws are different, it follows that biological theories and explanation must be different.
Hardy-Weinberg law: In an infinite, randomly mating population, and in the absence of mutation, immigration, emigration, and natural selection, gene frequencies and the distribution of genotypes remain constant from generation to generation.
Given the assumptions of the Hardy-Weinberg law, it follows from algebra alone that if we know the proportion of alleles in one generation, then we will know the proportion in the next generation. Hardy claimed to be embarrassed that this “law” should bear his name given that it is nothing more than a trivial mathematical deduction. This “law” is a priori, nonexperimental, and nonobservational. Rosenberg and McShea say:
… such definitionally necessary truths cannot report contingent causal connections, have no explanatory power, cannot support counter-factuals, and therefore cannot be laws, at least not of the sort we are familiar with in other natural sciences.
(While it is right that this is not a contingent causal connection, I’m not too certain about their latter two claims.)
R.A. Fisher’s sex ratio model is used to explain why most sexually reproducing species have a 1:1 female-to-male sex ratio. This model begins with some assumptions about the population, then derives from logic alone that the set ratio will be 1:1. This sex ratio model turns out to be a necessary truth.
As Rosenberg and McShea say:
All of these models are necessary truths. Their empirical core, which gives them a role in the explanation of the sex ratio of a species, is the claim that the species satisfies the assumptions of the model. Once we establish that the assumptions are satisfied, logic alone suffices to conclude that the model’s implications must be true.
In physics you have an empirical law where you fill in the initial conditions to get the result. In the mathematical models in biology, you don’t have empirical law. You have initial conditions, and the results follow from math alone. The empirical part comes from discovering whether a state of affairs satisfies those initial conditions.
(I’ve left talk of the principle of natural selection out of this because that deserves its own post.)
Rosenberg, Alex & McShea, Daniel. Philosophy of Biology.