There are two sorts of theories of semantics. By one sort of theory, if you find yourself at an international event or an airport lounge sharing a coffee with a stranger, and you and your fellow traveler come from opposite sides of the world, have no language in common, there are no interpreters around, and the other person’s language is not included in Google Translate, you want to say that you are committed to diversity and sustainability, you don’t need to make sure that your fellow traveller’s language has the basis for such ideas because semantics are essentially universal. They are based in a universal system of human reasoning and respect for life. They can always be translated somehow, because they are universal ideas. The framework here is strongly committed to this sort of universalist theory. By this theory, semantics falls outside the ‘learnability space’ and thus does not need to be learnt. So children do not normally have to be told that people and things have names, that entities can be referred to by deliberate acts, that some things are painful or dangerous, that there is such a thing as possibility, and that this is different from probability and from permanence.  There are difficult ideas, and there are things which can seem impossible. But remoteness and unfamiliarity are not ultimate barriers.

Perversely, the notion of a universal semantics was taken for granted by those who assumed that slaves could be ‘taught a lesson’ without a single word of explanation in a commonly shared language, as it is still thought that ‘human animals’ deserve to be punished. If they are not fully human, how can they understand or learn?

By the other sort of theory, with a deep history in human culture, semantics is relative, influenced or even determined by ways of life and accidents of what a language happens to include in its lexicon. A strong version of the second theory is unfairly known as ‘the Sapir-Whorf hypothesis’. Whorf got the idea by extraolating from Sapir’s thinking about how language and culture are intertwined. Sapir was a major scholar of various native American languages. And he deeply respected his informants. From one informant’s insight, Sapir developed the idea that there are properties in language which native speakers can detect, but are not overtly expressed. From this developed the modern notion of derivation, as by the G sound in English long, strong and young, only expressed overtly in longer, stronger, younger. By some understandings of the Sapir-Whorf hypothesis, the understandings of people in Stone Age cultures are different from those of people born in modern capital cities. There are ultimate barriers. But defeatism here is specious. The fact that explanation can be hard does not make it impossible. I choose to believe that the Sapir Whorf hypothesis is profoundly misleading.

One universal is the sense of touch and the consquential notion of a surface. Noam Chomsky (1995b) exampled the apparent universality of the notion of ‘nearness’. Obviously ‘near London’ and ‘near the cat’s nose’, involve different distances. But the notion of nearness excludes other frames of reference. If some bird or insect is flying a certain distance from the surface of a mountain, it might be said to be flying ‘near the mountain’. But if it flies into a cave on the mountain, it can no longer be said to be ‘near the mountain’, but only ‘in the cave’ because there is a nearer surface. Nearness denotes a degree of proximity to some relevant surface. It may be that it is related to the core notion of surface on which locative relations and prepositions seem to be defined. The objective world is partly defined by the relations that hold within it.

In, on, and under do not refer to absolute and invariant directions, but to the notion of surfaces. On is widely supposed to involve a notion of verticality, where one entity is above another, closer to the sky, further from the ground. But a barnacle makes its home ‘on the bottom of a ship’, a fly stops ‘on the ceiling’, reversing the polarity. Rather than denoting verticality, on denotes contact with some surface. Similarly, under is widely thought to involve the polar opposite of on. But a submarine travels ‘under water’, meaning that it travels under the surface of the water. Dishes ‘in the sink’ or someone ‘in the bath’ are not said to be ‘under water’, but ‘in the water’ because there is a more immediate enclosure, as in the case of the bird who has flown into a cave. A duck swimming on the surface of a pond is said to be ‘in the water’ as long as it is on the surface, but ‘under water’ as soon as it dives. In thus denotes some degree of enclosure. What appears to be universal here is the frame of reference rather than a spatial relation.

Roles in conversation change from one item of dialogue to the next. In “I didn’t say that” and “You most certainly did”, I and you may refer to the same individual. None of these things vary from language to language. They do not or should not need to be taught. If any of these things do need to be taught, this is a different sort of therapeutic or educational enterprise from work on phonology or syntax.

It is sometimes suggested that meaning is essentially categorial, that living things can be caregoristed as animate or inanimate, and so on, until one gets to particular species or social groups. But this breaks down at the point when the meaning denotes an attitude in the mind of the speaker. Describing a man as a bloke or a geezer, or what he is doing as ambling or wittering, suggests a lack of respect. Attitude cuts across categorisation.

Semantics is the dark matter of linguistics. It has to be assumed. But the way it works is up for grabs.