Faith & Science
Seeking an Official Definition of “Randomness”: A Reply to Jay Richards
I’d like to thank Jay Richards for his interesting and insightful review of my book Where the Conflict Really Lies. One relatively minor matter: it wasn’t my intention to show that all of current science is compatible with Christian theism. I thought some parts (current physics, current biology) are, and some parts aren’t: various theories proposed in evolutionary psychology, for example. What I wanted to show about these latter theories is that even if they constitute good science, they don’t provide defeaters for the Christian beliefs with which they are incompatible.
For the most part, however, Jay and I are on the same page. Still, we do have something of a disagreement, or near-disagreement, which shows up in the third installment of his review.
The question is about the scientific theory of evolution. This theory has several parts: universal common ancestry, descent with modification, and what I called, perhaps infelicitously, Darwinism. (It also includes the idea that the world is very old, and that human beings have simian ancestors.) I proposed to use the term is a specific and circumscribed way: as a name for a scientific theory, as opposed to the scientific theory plus comments or glosses thereon. This theory is a theory about the process that drives descent with modification: it is the claim that what drives that process is natural selection working on genetic variation, where the must popular candidate is random genetic mutation.
I took that theory to be compatible with the thought that the process of evolution is guided by God. I took it that the theory as such doesn’t entail that the process is unguided, or a matter of chance. Of course the theory also doesn’t include the claim that this process is guided by God; it doesn’t as such address the question of guidedness at all. It leaves that whole question open.
Jay disagrees. He believes this scientific theory as such, “Darwinism” as I called it, does include unguidedness. He thinks Darwinism, that scientific theory, includes as a part or as an entailment, that the process is not guided by God or anyone else.
Now one question here, and it’s an interesting one, is this: how do we tell precisely what a scientific theory is? How do we tell exactly what the scientific theory of evolution is? There is no axiomatized presentation of the theory emblazoned on the walls of the American Academy for the Advancement of Science. Who gets to say what the theory is? I was proceeding as follows. First, as far as I can see, the only place where this question of guideness vs. unguidedness arises is with respect to the claim that what drives the process of descent with modification is natural selection operating on genetic mutations, where the genetic mutations are said to be random. The whole question here turns on the meaning of that term “random.” Clearly the term has many different meanings: as Jay points out, there are several common meanings to be found in dictionaries. Some (but not all) of those meanings are incompatible with guidance; but this doesn’t settle the issue. There is also the meaning of the term as used in mathematics (“random” sequences, for example), which is quite different. If you are asked for the meaning of “random” in mathematics, you won’t do well to quote the dictionary, even though there will be an analogical relation between the mathematical sense and one or another of the everyday senses of the term. The same goes for the biological meaning of the term: you won’t find that by looking in the dictionary.
My thought was that the way to proceed is to see what official or semi-official definitions of the term “random” are offered by leaders in the field when they are trying to give a strict definition of the term, not when they are speaking more generally about evolution. The ones I settled on were by Ernst Mayr, perhaps the most distinguished evolutionary biologist of the 20th century, and Elliott Sober, perhaps the most distinguished contemporary philosopher of biology. According to Mayr, “When it is said that mutation or variation is random, the statement simply means that there is no correlation between the production of new genotypes and the adaptational needs of an organism in a given environment.”1 And according to Sober, “. . . there is no physical mechanism (either inside organisms or outside of them) that detects which mutations would be beneficial and causes those mutations to occur.”2 Mayr was not a theist, and neither is Sober; furthermore neither has any particular interest in arguing that theistic belief is compatible with Darwinism. Richards complains that these definitions are “very narrow”, and he doesn’t think they are representative. But I wasn’t able to find any official definitions of the term that point in another direction. Sahotra Sarkar, another anti-theist, puts it as follows:
The critical point is that the theory of natural selection is neutral about whether the origin of variation is blind (undirected) or not; all natural selection requires are (i) the existence of variation (whether or not these arose blindly), (ii) fitness differences between the variants, and (iii) inheritance of the variation (Lewontin 1970). Given these assumptions, the fitness differences lead to directional selection resulting in increased adaptation between an organism and its environment.
Michael Ruse, also no theist, defines “random” as follows: “The ‘raw stuff’ of biological evolution (i.e. mutations) is random, in the sense that it does not occur according to need.”3 Ruse lists a number of other authors who, he suggests, say the same: L.J. Cohen, P. Thagard, R. Amundson, C. Kary, and C. Hookway. Jerry Coyne, hardly a theist or someone with an interest in arguing for consistency between evolution and theism, says the very same thing.
As I say, I haven’t been able to find any official or semi-official definitions of “random” that point in a different direction.
If we take a closer look at Mayr’s definition, which seems to be a more formal version of the definition offered by Ruse et al., it comes to something like this. Take the class of all the mutations that have occurred: it is not the case that all or most of them have filled adaptional needs of the organisms to which those mutations have accrued. This is clearly compatible with these mutations having been caused by God in order to accomplish his purposes.
Richards also says, “In fact, even if all Darwinists adhered to Sober’s precise definition, most would assume that Sober is offering a distinction without a difference, since physical mechanisms are the only mechanisms that exist.”
Here, first, I wonder whether Richards is sticking to the definition of “Darwinism” (and by extension, “Darwinists”) that I gave: “Darwinism,” the way I’m using it, is no more than the claim that what drives descent with modification is natural selection operating on random genetic mutation. And second, even if most would make the assumption Richards mentions, that wouldn’t show anything about the scientific theory as such. Even if most physicists assumed that God promulgates the natural laws for the universe, it wouldn’t follow that the statement of physical laws (Newton’s Laws, e.g.) involve that assumption.
For Darwin and most Darwinists, in fact, random doesn’t just mean uncorrelated to a physical mechanism that works in favor of organisms. Random mutations have some or another cause, to be sure, but the cause, in the Darwinian view, is blind, unguided, and purposeless.
Here again I wonder whether Richards isn’t using “Darwinist” more broadly than I was. On my definition, some of the people at BioLogos are Darwinists; but they don’t all think the cause has to be blind, unguided, and purposeless. And in any event, once more even if Richards is right here, it doesn’t show anything about the scientific theory. Again, suppose nearly all physicists thought the laws of quantum mechanics were established by God: it wouldn’t follow that according to quantum mechanics the laws of quantum mechanics were established by God
At the same time, he [i.e., me] wants to show that there’s nothing in theistic religion that conflicts with “science.” This goal is nowhere more difficult than when he deals with the reigning theory of biological evolution, depending as it does on the Darwinian process of natural selection and “random” mutations. As a result, Plantinga must find a “scientifically reputable” definition (my phrase) of the word “random.”
This is misleading: (1) I don’t want to show that there’s nothing in theistic religion that conflicts with science: as I said, a lot of evolutionary psychology and scientific study of religion does conflict with theistic religion. And (2) it isn’t as if I assumed in advance that evolution and theistic religion are compatible; I’d be perfectly content if they turned out not to be. Rather, I wanted to look to see whether they were. It seemed to me that the locus of incompatibility, if there were incompatibility, would be with the meaning of the term “random”; so I looked for definitions by the most distinguished writers on the subject (Mayr and Sober but also those others I mentioned above).
Richards also speaks of “Sober’s highly restrained definition of ‘random'”; as I say, though, I couldn’t find any official definitions of the term that pointed in a different direction. Richards also says:
The highly specialized definition of “random,” for instance, which is compatible with God’s guidance, just is not the Darwinian meaning of the word — despite the fact that some Darwinian philosophers and biologists, in moments of caution or strategic cleverness — have offered such a definition.
Again, this seems to me misleading. First, once more, I don’t think Richards is using “Darwinian” the way I did. I think he is instead referring to a group of people who accept the scientific theory but also believe that there is no design or guidance in the process of evolution. Otherwise I can’t see what he would be meaning by “the Darwinian meaning of the word.” And the philosophers and biologists whose definitions I cited weren’t, I think, being cautious or strategically clever. They aren’t theists and don’t have any stake in the thought that theism and evolution are compatible. They’re just saying how the term “random” in the theory is to be understood.
Richards is certainly right in saying (1) that lots of evolutionists and probably most biologists believe there is no design in nature, and talk and write as if evolution shows this to be true, and (2) the current cultural significance of evolution lies in that direction. But why shouldn’t IDer’s point out that the official definition of randomness is such that Darwinism (taken my way) is perfectly compatible with theistic guidance, or (2) if they don’t think I’m right about the official definition, why shouldn’t IDer’s point out that there is this alternative (as they see it) way of taking Darwinism, and that so taken it isn’t incompatible with Divine Design?
(1) Towards a New Philosophy of Biology: Observations of an Evolutionist (Cambridge: Harvard University Press, 1988), p. 98.
(2) “Evolution Without Metaphysics?” in J. Kvanvig (ed.), Oxford Studies in Philosophy of Religion, volume 3.
(3) Philosophy of Biology Today (Albany: State University of New York Press, 1988), p. 75.