The Essence of Language: How Words Were Born and Evolved (Chuko Shinsho, by Mutsumi Imai and Kimi Akita)

This is a book from the 2024 Shinsho Awards.
I was finally able to read it, albeit a bit late.

※ I am a participant in the Amazon Associates program.

Mutsumi Imai has a new book out recently, and I'd like to read that one too.

"The Essence of Language."It is a very grand theme.
I think it takes a lot of courage to delve into this, as it is almost too grand.
If one were to write something on the theme of "the essence of language," there could be many different approaches and contents. Is it a linguistic meaning, a historical meaning, a philosophical meaning, or perhaps a communication theory meaning...

I believe the greatest feature of this book lies in how it approaches this grand theme.

To be honest, I had the impression that it was such a unique way to start. Huh? You're entering into "the essence of language" from there? It was that kind of unexpected approach.

Perhaps it's because I'm a bit rigid, but when someone tries to talk about "the essence of language," I tend to expect a writing style that first discusses what "language" is, defines it, and then discusses the important elements for language—the "essence"—within that definition.

This book takes an approach that goes against such rigid expectations.

And that approach is... it is, quite simply,

"Onomatopoeia."

This book cuts into the forest of "the essence of language" starting from "onomatopoeia."

Onomatopoeia refers to sound symbolism, like "beta-beta" (sticky), "wan-wan" (woof-woof), or "goro-goro" (rumbling/rolling).

Cutting into "the essence of language" from here.
At this point, it's already full of a sense of wonder.

Perhaps some of those who have already read this book might have had the impression that "it was different from what I first thought" after reading it. I think that's probably because they felt it was approaching "the essence of language" from a completely different angle than they had initially expected, making it a bit hard to follow.

I think it is actually that much of an unexpected way to enter the subject.

Linguists define onomatopoeia as "words that have a characteristic form that captures sensory images and can be newly created." In other words, onomatopoeia has an iconic quality that captures a part of things in an iconic way.

For example, "hara-hara," "bara-bara," and "para-para" have different sensory images. You can sort of feel the difference between the "h," "b," and "p," right?

This "capturing of sensory images by making full use of iconicity" is the key to approaching the "essence of language" in this book.

On the other hand, if you ask whether onomatopoeia is language, it falls into the category of "language" rather than non-linguistic sounds. This is language when compared to non-linguistic sounds like whistling or throat-clearing.
In other words, perhaps onomatopoeia possesses both linguistic and physical properties, or is positioned somewhere between the two.

The reasoning process that attempts to approach the "essence of language" using this as a starting point.

In other words, while language is a system of abstract symbols, its birth—that is, the trigger for humans to start using language—must have possessed some kind of physical property. If so, by analyzing onomatopoeia, might we be able to approach the "essence of language"?

I think this is the highlight of the early part. I found this perspective interesting.

＊

Now, how to handle onomatopoeia.

The middle part focuses on the process of language acquisition in children.

Onomatopoeia is sensory and physical. It is not rare for adults to be impressed by the novel onomatopoeia used by children.

However, language does not end with mere "physicality." Words eventually leave physicality behind and begin to have their own unique meanings.
Children start connecting meanings and sounds on their own. In fact, children sometimes make cute errors of generalization.

One such example was given.

Children generalize the words and meanings they have acquired. Here, eating a mandarin orange is also generalized as "acheche."

That process is, in a sense, an arbitrary generalization made after the child has thought about it. Conversely, what is happening here is that through inference, the child is moving away from a world of only onomatopoeia, the level of abstraction is increasing, and at the same time, the iconicity of the words is fading.
In this way, what started as language acquisition that only connected many sensory images to sounds gradually organizes and connects the relationship between word forms (sounds) and meanings, and as a result, the number and variety of words known increases.

And actually, it doesn't end here. Next, they start to collect similar words from among the words that have been arbitrarily generalized and abstracted, and begin to systematize them.

What happens then?

The book says that another iconicity is born again.

For example, as a result of the systematization of words, images such as "h" being light and "b" being heavy emerge. It seems that this does not deviate much even when looking at other languages globally. Sensory images obtained physically are symbolized, generalized, and another iconicity that does not go through individual sensory images is born.

This is a different iconicity from the iconicity that captures individual sensations based on onomatopoeia. It might be said to be an iconicity that only emerges for the first time by collecting words and finding commonalities.

The book organizes this cycle as "Primary iconicity → Arbitrariness → Systematization → Secondary iconicity".

Now, having organized it this far, how does the book approach the "essence of language" from here...? Why was iconicity divided into two? Why was this cycle proposed...?

You might be getting curious... but since this has become long, let's leave it at this (laughs).

Here, I would like to offer a few hints to approach this mystery, and then leave the rest for you to discover by reading this book.

One is the "symbol grounding problem." This is apparently a known issue in artificial intelligence. It is described as the problem that "in order to understand the true meaning of words, one must have physical experience with the object as a whole." For example, if an AI were fed the symbol "strawberries are sweet, sour, and delicious," it might conclude through symbol analysis that the taste of a pineapple is also the taste of a strawberry. Can an AI that has been fed a massive amount of symbolic information truly be said to know what a strawberry is?

Another is "abductive reasoning."
In this book, it is organized as the third type of reasoning, following induction and deduction.
It is often used in the formation of scientific hypotheses.It is like when Newton saw an apple fall from a tree and formulated the hypothesis of the law of universal gravitation.

The characteristic of this reasoning is that it is reasoning for hypothesis formation to discover general laws that are not yet understood. Therefore, it means there is a possibility that it could be wrong (deeply meaningful).

By the time you finish reading, you will realize that this book itself is a grand abductive inference.

With onomatopoeia, the two types of iconicity, the symbol grounding problem, and abductive reasoning as hints, how does this book approach the "essence of language"!? If you are curious, please give it a read.

I wonder if the latest book I mentioned at the beginning, "Loss of Academic Ability," is a new publication based on this reasoning; it makes me want to read it again.

With that said, thinking that the number of books to read by the end of the year has increased again, I wish you a "great day today."