Two Ancestral Languages

Two Ancestral Languages¹

“Tell me a story, Father. Tell me of the world before I was born. But first, repeat the story of your own father’s life and the world in which he lived, as he told it to you in every detail. And before that, your grandfather’s life and world, as he detailed it to your father. And before that . . . great great great great great grandfather’s life story in his world, as he told it to great great great great grandfather. How did they live all those centuries ago? What was the world like?” I want to know everything about those pasts, so I can pass the information on to my own children, and nothing will be forgotten.

Theoretically, spoken language has the narrative power to preserve, from parent to child down an indefinite succession of generations, a full and detailed record of ancestral history. I don’t need to tell you it doesn’t work like that. It is derisory how little information trickles through. And that’s no bad thing. The sheer volume of information that would accumulate with each generation would overwhelm us. Of my eight great grandparents, I know either nothing or little more than their names. That is sad but inevitable. Even written records, letters, diaries, chronicles, are so fragmentary that historians can’t agree on what happened to a whole country, let alone how to interpret it. Yet language potentially could have transmitted a detailed record – everything each generation might say to the next – across indefinite generations. 

Language is the defining glory of the human species and a wonder of the world. Such forerunners as the bee dance, or whale or bird song, hardly begin to compare. It is language above all that grants to our species the cumulative power of cultural evolution. Where science, technology and the arts are concerned, each generation builds on the knowledge and achievements of its predecessors. Like Newton we stand on the shoulders of giants², none loftier than Newton himself but commonly on the lesser shoulders of ordinary individuals who, in turn, gained height from others. Cumulatively we rise to where our vision outreaches a lone individual’s horizon, toward the Event Horizon of the very cosmos itself, and back to the hot birth of time. And yet we know almost nothing of what our great grandparents could have told us.

The collective power endowed by language is captured in the legend of the Tower of Babel, where it threatened even God himself. “Go to,” God said, “Let us go down, and there confound their language, that they may not understand one another’s speech.” And lo, as a result of quasi-evolutionary divergence, we speak many thousands of mutually unintelligible tongues, the exact number undefined, mostly because dialect grades into language so we can’t decide where one language ends and the next begins. This smearing out is seen both across geographical space (think English worldwide) and through historical time as languages evolve through the centuries (try reading Chaucer). Nevertheless, because language is partly digital, chunked into the discrete semantic units that we call words, it is capable of great fidelity of transmission, especially in written forms. Yet, despite being thus capable, precious little verbal information from our dead ancestors filters down. 

But there is another language, also digital, of far greater precision and fidelity of transmission, which does preserve ancestral information, not through a paltry few generations but through hundreds of millions; a parallel language, which is not a human monopoly but is shared by all life: all animals, fungi, plants, bacteria, and archaea; a truly universal language that doesn’t suffer the confusion of dialect drift or trans-generational decay. Nor does it fall victim to the cumulative inflation of my opening fantasy, because it does not grow with each generation. As every schoolchild knows, acquired characteristics are not inherited. The genetic database changes, not cumulatively but by subtraction and addition approximately keeping pace with each other.

This other language is so relentlessly universal that a Babel-fearing deity should have nipped it in the bud long since. Written in an alphabet of four letters, the universal dictionary of DNA has only 64 words, each three letters long. Synonyms prune the effective dictionary from 64 words to only twenty codons plus a punctuation mark. Typically, an amino acid is coded by a few synonymous triplets. This synonymy means the code is technically “degenerate”, not “redundant” as is often mistakenly said in biology textbooks³. The 20 semantic meanings are amino acids, strung together into protein chains. Each protein chain can be thought of as a sentence, but in an impoverished sense that doesn’t do justice to the analogy with language, and it is here that the analogy runs out of steam. The complications of embryology disclose a richer semantics. “Sentence” and “paragraph” are unpersuasive here, but a higher meaning emerges at the end of long, interacting causal chains, in the form of “phenotype” – stripes and colours, hearts and lungs, backbones and behaviour: plus what I have called the extended phenotype.

It is only within my lifetime that science has been able to read the language of DNA in its direct, digital form. But we have long been able to read its indirect manifestations, its phenotypic effects. We read them in the shapes and colours of animals and plants, fungi and microbes, in the elaborate courtship displays and other rituals of birds and spiders, in the shattering complexity of cells and nerve nets, the arresting colours and scents of flowers, the software that runs in brains and ganglia, the hardware of bone, chitin, and wood. My next book (2024), The Genetic Book of the Dead, will be mainly concerned with what we can read from these indirect manifestations, although biologists of the future will not be so limited, and will read the genetic book as well.

Unlike English or Kikuyu, German or Cantonese, the language of DNA is not used for inter-individual communication⁴. It is used for transgenerational communication, and of course for directing the course of embryonic development within each generation. At first sight, the transgenerational communication is not of the kind with which I began this introduction, of father regaling child with life story. Acquired characteristics, as I now repeat, are not inherited. And yet there is a sense in which DNA encodes a rich description of ancestral worlds, even if it is nuanced and hard to decipher. My thesis will be that the world of a living organism, more specifically the successive worlds in which its ancestors lived, is described by – can be read in – its body, its behaviour, and its DNA. Your body, and its genome, can be read as a book, a history book describing the successive worlds in which your ancestors survived. This is the meaning I intend by The Genetic Book of the Dead. 

Richard Dawkins

1. Originally written as the Prologue to my next book, The Genetic Book of the Dead, to be published by Head of Zeus, London, in autumn 2024. I decided to remove the Prologue from the book, preferring to publish it here instead.

2. Newton’s famous phrase occurs in his 1675 letter to Robert Hooke, written in terms whose generosity belies their later feud, and belies the rumour that he was being sarcastic at the expense of Hooke’s kyphosis of the spine. Newton may have been quoting Bernard of Chartres (c .1130), whose words fit my purpose more fully even than Newton’s: “We are like dwarfs on the shoulders of giants, so that we can see more than they, and things at a greater distance, not by virtue of any sharpness of sight on our part, or any physical distinction, but because we are carried high and raised up by their giant size”.

3. Redundant repetition guards against error. Like when a bank cheque says “One hundred pounds . . . . £100”, mentioning the quantity twice for safety. That’s redundancy. The 64-word DNA dictionary has synonyms, but that’s a different matter entirely. Synonyms don’t constitute redundancy. There are two alternative “words” for the amino acid Asparagine, namely AAC and AAT, but the genetic code doesn’t redundantly say “AAC AAT” to be doubly sure of making asparagine. It says either AAC or AAT. Textbooks are wrong to say synonyms make the code redundant. The correct technical term is degenerate. There may be redundancy of another kind in the genetic code, but it does not reside in the fact that any one amino acid can be coded by more than one RNA triplet.

4. But it could be. It’s quite rich enough. You could easily devise a code whereby each of the 64 DNA triplets represented a letter of the alphabet, a number, or a punctuation mark. You could then write a message, say the complete text of this book, insert it into the genome of a rhinovirus, infect yourself with the virus, sneeze it out in a crowded airport, and spread the book around the world in an epidemic. Molecular geneticists anywhere in the world could then sequence the genome and print out the book. Of course there are better ways to increase sales.