The Kojiki Compilers' Struggle Between Sound and Script Returns in the Reading of Genes｜COMT

Last week I talked a little about genes, but it is hard to get a concrete feel for them. So this time I will focus on a single gene. Before that, though, one interesting story.

When we think of Japan’s oldest surviving texts, we think of the Kojiki and the Nihon Shoki, compiled in the early eighth century. Our genes, however, are a text older than the Kojiki, which was completed in 712. They are, in a sense, a chronicle without a compiler, lying asleep inside each one of our bodies.

That is precisely why the people who compile genetic data matter so much. The project that produced the Kojiki began because the various clans each held to their own traditions, and there was no bringing them into order. Emperor Tenmu, intent on a system for a new age, issued an edict, and what began as what we would now call a national project moved toward completion with the powerful backing of Fujiwara no Fuhito.

The leader of the Kojiki project has long been said to be O no Yasumaro, who wrote its preface, but it has now been shown that someone in a later age wrote the preface under his name. What matters is that turning the language of that ancient age into writing, and conveying it to readers, involved difficulties beyond imagining.

The Chinese characters in use at the time were tools for writing Chinese, so every one of them already carried meanings from the continent. Suppose you wanted to write “ama,” a sound in the old Japanese language. There was no character that stood for the sound “ama” alone. Write 天, the character for heaven, and the meaning comes across, but nothing settles whether the reader will say “ama” or “ten.”

If instead you picked up only the sound and wrote 阿麻, you could fix the reading, but the meaning vanished from the written form, and the number of characters more than doubled. When Yasumaro wrote in the preface that the task was “difficult,” this was the dilemma he meant. Take the meaning and the sound disappears, while taking the sound makes the meaning disappear and the text grow long. Neither way could work on its own. How to seal the sounds of Japan into writing was a hard problem that lay across the path from the very start.

So the compilers did something quite awkward for its time, mixing both methods within a single sentence. Wherever the reading might be lost, they went around adding notes one by one, saying in effect, “read this 天 as ama.” The sheer number of those notes is the flip side of the fact that the text could not have stood without them, and in a sense it was a groundbreaking method.

The songs, on the other hand, they wrote entirely in sound. In a song, the sound itself is the substance, and once you replace it with meaning it is no longer a song. Knowing full well the text would grow longer, they gave sound priority there. To translate this formless culture of sound into the visible form of written characters is exactly where a compiler’s gifts are put to the test, and it was a great undertaking whose effects would reach across the following centuries, even a thousand years.

The Nihon Shoki, incidentally, made the opposite choice. It was written in proper classical Chinese, and in exchange it largely abandoned sound. As an official history meant to be shown to the outside world, it chose form. In the same age, then, a book that protected sound and a book that protected form were made side by side.

That same sensibility is arising again today, in the translation and compilation of genes. Because genetic research has long been carried out by Western standards, it is based in English, much like programming languages. Yet just as the ancient clans each had their own traditions and interpretations, the meaning and interpretation of genetic data change completely depending on where a given population belongs.

In other words, to take the results of a genetic analysis and read further into them is to take on the very problem that the Kojiki’s compilers ran into. That is what makes it so interesting.

Until now, however, the only grammars in circulation for reading this text have been those written in the West, and the grammar of genetic information is almost without exception based on white populations. A script and grammar created for white people have simply had to be forced onto non-white people. This is a serious matter, one that can be seen as an extension of colonialism. Yet when the grammar differs, the same letters take on a different meaning. Let us touch on that, just once.

The gene I want to look at is COMT, catechol-O-methyltransferase. Put simply, it is the gene that produces an enzyme that breaks down neurotransmitters such as dopamine, adrenaline, and noradrenaline.

What matters is whether the 158th amino acid of that enzyme is valine (Val) or methionine (Met). This difference “in a single spot” changes how actively the enzyme works by a factor of three to four. There are three types, Val/Val, Val/Met, and Met/Met.

To sketch their traits roughly, the Val/Val type metabolizes neurotransmitters quickly, so dopamine in the prefrontal cortex is cleared away fast. Because of this, in ordinary, peaceful times, their concentration tends to scatter and they find it hard to show what they can do, yet under stress they display remarkable strength. Their strong drive toward novelty, always heading on into unexplored territory, comes at a price, and they struggle to focus on one thing.

Met/Met is the reverse. Because dopamine lingers, these people lose their footing under stress, but in a quiet place they have a contemplative power that lets them sink deeply into a single thing.

Val/Met is an intermediate type that carries something of both. On the basis of these traits, David Goldman, a geneticist at the US National Institutes of Health (NIH), distinguished the two types as the “warrior” and the “worrier.” This is why Val/Val is sometimes called the warrior gene, and Met/Met the worrier gene. Curiosity and absorption, exploration and lingering. It is one of the switches at the root of human temperament.

Interestingly, in the prefrontal cortex, which governs thinking, the clearing of dopamine is known to depend heavily on COMT. This gene, in other words, does not simply shape temperament. It is one of the important genes that bear on “the way we think” itself. And even in COMT alone, there are decisive differences between populations.

Next week, I will read into this.