- Genes – the language of God 0: Preface
- Genes – the language of God 1: Genes as Language
- Genes – the language of God 2: Other popular gene myths and metaphors
- Genes – the language of God 3: Why genes aren’t information
- Genes – the language of God 4: Why genes aren’t a language
- Genes – the language of God 5: God and genes
- Genes – the language of God 6: Theological implications
If we are to ask whether genes are God’s (or anyone else’s) language, we better first ask what is a language. Languages are divided into natural and formal languages. Human language is a natural language because it evolved haphazardly, and its utility is the result of a process not unlike natural selection in biology. So the word “dog” in English applies to canines more or less (does it cover coyotes? African wild dogs? Jackals?) because of a long association between the word and dogs in the minds of English speakers and listeners. The word “dog” is about dogs (philosophers say, it “refers to” dogs; and in philosophy, when a word has quote marks, it is being mentioned as a term; when it does not it is being used to refer to the things it refers to).
Now, natural languages have grammars, but they also have exceptions; they are messy and inconsistent. Formal languages (like a well-defined programming language, or mathematics) are consistent and not messy (though they can be complex). We might ask then whether genes are a natural language or a formal language. The genetic code evolved, and it evolved so that each triplet code results in a particular amino acid residue, so it might be a natural language. On the other hand, it is highly structured and insistent, so perhaps it is a formal language.
Arguments against the naturalness of DNA as a language sometimes rely upon the fact that common aspects of natural languages are not found in DNA:proteins relations. Here are two:
All known human languages exhibit something known as “Zipf’s Law”. This is a statistical feature of many kinds of large data sets (i.e., of measurements of populations), which, when mapped on a log-log graph, display an inverse linear distribution:
The relationship between linguistic sequences of a certain length and their frequency should be this sort of inverse linear kind. However, when DNA sequences are analysed, they do not exhibit this sort of relation (Tsonis et al. 1997). That doesn’t mean DNA isn’t a language, but it does undercut the claim out of the box.
Another argument is the problem of displacement: human languages involve reference to things that are not present, such as an absent friend. But DNA does not “refer” to anything; instead, it produces the RNA products.
Now there is a kind of language act that makes things happen: it is called a “performative” act. An example is the sentence: “You are under arrest”, by a policeman, which makes the person be under arrest, or “I now pronounce you husband and wife” when said in the right circumstances by a marriage celebrant. The act of saying it makes it so. But performative acts are not the whole, or even a large part, of language. We tend, at least in the western tradition, to make words have power. Consider the opening of the Gospel of John:
In the beginning was the Word (logos) and the Word was with God, and the Word was God. He was with God in the beginning. Through him all things were made; without him nothing was made that has been made.
In the older Jewish tradition, such a role was played by Wisdom, a kind of causal instrument of God. And God’s own words had power, as the opening of Genesis that John’s gospel paralleled indicated. So the temptation is to think that in science, words (here, the “words” of genes) also have power.
I think this is a mistake caused by a false transference of human (and potentially divine) powers to the non-human world, which is usually referred to as “anthropomorphism” or projecting human properties to nonhuman phenomena in the world. It generally has been regarded as an error in science for four centuries (but not always by philosophers or religious believers).
In science, causation is about physical interactions, not semantic interactions (except when a semantic processor such as the one between the ears of any normal human being over the age of 4 is processing words as received sounds, etc.). DNA acts as a template because it forms weak bonds between the nucleotide sequences and the RNA monomers in the polymerase.
So perhaps what we mean by “X is a language God uses to do Y” is that God uses regular causes to achieve some or all of his aims. This is not inconsistent with any science, of course, since science works by positing and testing regularities of phenomena, and calling them causes. However, it has an unwanted side effect: any regular cause of any kind at all becomes “the language God uses to do Y”. So chemical bonds are a language of God, as are the laws of physics, and so on. This may be acceptable to many, but it seems to me that in the end the idea of it being a language becomes unnecessary: it is enough to talk about causes (and we will in the next post, do precisely that). In the event that DNA is a language because it is a regularity in the world, there would seem to be an indefinitely large number of such languages in the universe. [And this might encourage those who see God as a geometer or mathematician to suggest that the language of God is numbers, not genes. This is a widely held view.]
So on balance, I think that referring to DNA as a language is generally fairly harmless (unlike the use of “information”, which is a much more abstract property, and more misleading), but that it is not something that makes any claims with content. It is an analogy only, and doesn’t illuminate things to any real degree. One has to ask, what is gained by calling DNA a language?
One advantage, shared also in calling genes information, is that it sets up a “problem” that can be “solved” by divine action or intention. For example, one of the main kinds of advocates for genes as linguistic things is the intelligent design movement. If genes are meaningful, or the result of some intent, then that implies that one needs to have a deity (or something very like it) to give that meaning. If genes are not informational or linguistic things, though, but at best only something that can be analogised to language, then that “solution” is no solution at all.
I do not wish to suggest that everyone who uses this analogy intends for there to be a Designer or Speaker of genes; obviously many do not (including Dawkins). But it seems to me that we give away too much territory to the Intelligent Design folk by accepting in the first instance that genes are informational or linguistic. In the next post I will argue (in dialogue with Stephen Ames) that even theists need not make this analogy carry too much weight, and that they can as easily accept unintentional processes as they can any other kind of scientifically investigable process.
Tsonis, A., Elsner, J., & Tsonis, P. (1997). Is DNA a language? J Theor Biol, 184, 25–29.