‘Through the Language Glass’ is Guy Deutscher’s second book. His first, ‘The Unfolding of Language’ argued that linguistic change is cyclical, that languages go through stages of decay and reconstruction. ‘Through the Language Glass’ tackles the Sapir-Whorf hypothesis of linguistic relativity, which proposes that our native language limits the way we think. The hypothesis is mostly discredited, and Deutscher doesn’t aspire to rehabilitate it, but he does try to salvage some pieces from the wreckage. He doesn’t suggest that the language we speak limits our potential for logical reasoning, but he shows how it influences the way different people might see the world in three specific areas: colour, spatial awareness and gender. Part of his argument is based on the fact that not only do different languages allow us to do certain things, they make us do certain things. In English, we are obliged to say whether the owner of something is male or female. We have to say either ‘his house’ or ‘her house’. In French, by contrast, speakers have to assign gender not to the owner, but to the object. It can only be ‘sa maison’. It would do an injustice to the book to try to summarise the argument further, but I’d just like to mention a fascinating point (at least, to me) from each of the three areas he considers.
First, colour. Deutscher starts by examining a work by William Ewart Gladstone, a nineteenth century British Prime Minister, on Homer. (No, I didn’t know he did, either.) Gladstone was struck by the limited range of colour words in Homer, but Deutscher moves on from this to show that different languages divide up the colour spectrum differently. The Japanese word ‘ao’ once covered green and blue, but has now come to be applied mostly to blue shades and green is now mostly called ‘midori’. The ‘go’ light of the first traffic lights in Japan was called ‘ao’, but over time that didn’t seem an adequate description, given its developing use as a word meaning blue. The obvious thing to do might then have been to change the name of the colour of the ‘go’ light to ‘midori’, but, instead, the Japanese changed the colour of the light to match the name. And that, children, is why the ‘go’ light on Japanese traffic lights has a bluish tint.
Next, spatial awareness. Guugu Yimithirr was the language spoken by people on the north-east coast of Australia. Speakers of European and other languages will mostly locate people and things according to whether they are to the left or right or behind or in front of the speaker. Only in special circumstances will they use compass points. Not so the speakers of Guugu Yimithirr. They would always say that someone or something was to the north, east, south or west of someone or something else and wherever they were they will always knew how to do so.
Finally, gender. Deutscher articulates well a possible explanation for the origin of grammatical gender which I have been feeling my way towards myself. He gives the example of ‘aeroplane’, which in the Australian language Gurr-goni belongs to the vegetable gender. How could that possibly be? Well, the vegetable gender, he hypothesizes, would first have been extended to plants more generally, including trees. Then the wood from trees would have been included in the gender. Then things made from wood, such as a canoe. But what is a canoe? It’s a means of transport, so what more natural than that the vegetable gender should be extended to include all means of transport. And what is an aeroplane? Yeh, got it.
There have been a number of ingenious experiments to show that when an inanimate object has been assigned to male or female gender, speakers of languages that have grammatical gender will associate the object with male or female characteristics, not because such characteristics are innate in the object, but simply because the words themselves are grammatically one or the other.