Language and Meaning

Many years ago someone observed to me that until after the Second World War language had been presumed to be transparent. That is, that words meant what they said and said what they mean. As with most generalisations this was not quite the case, but as a broad brush it will do.

After the Second World War two things happened. The first was a degree of technological innovation, the start of digital communications. This led to the development of communications theory, an effort led by Claude Shannon. This rang a bell with me, because many years before the language lecture I am referring to, I did a course on communication physics, and struggled with an object called Shannon’s theory. If you read Shannon’s original papers (in Physics Review) they are quite easy to read, but the concepts are a bit mind bending.

Shannon observed that much everyday language is redundant. That is, much of what we say or write is unnecessary to communicate what we mean. If you work it out, I seem to recall that up to 80% of a sentence in English is redundant. Y__ c_n s__ wh_t I m__n _n t__s se____ce. It might well slow you down, working out the content of the missing letters, but the context and rules of grammar actually give you most of the missing letters.

We can, therefore, produce an ‘ideal’ language of, say zeros and ones, with no redundancy and be able to say exactly what we mean. The information content is maximised, the redundancy is zero. However, problems arise when you use a ‘lossy’ medium that is one with noise or dispersion in it. Then the difference between zeros and ones can be lost and errors occur in the message received.

In order to combat the errors, checking bits are added. Thus, for each batch of ones and zeros, an extra one or zero is added to give the sum of the preceding digits (we are in binary here, so 0+1 = 1 and 1+1 = 0) so that errors in the batch can be detected. This works, of course, but adds a degree of redundancy to our otherwise ideal language.

The problems multiply, of course, as distance and noise increase, and more sophisticated error checking is needed. However, this is the basis of our modern communication networks – even in a computer network redundancy is rife because it has to be to spot errors.

The other thing that happened was what is usually known in the history of philosophy as ‘the turn to language’. This was mainly due to the influence of Wittgenstein’s late work, Philosophical Investigations (Wittgenstein, L., Philosophical Investigations (Oxford: Blackwell, 2001 [1953])). In this book (which is a difficult and unusual read, by the way) Wittgenstein notes the variety of uses of language, not all of which are giving information.

The wide variety of uses of language Wittgenstein lists include giving and acting on orders, reporting an event, making up a story and reading it, translating, requesting, thanking, cursing, greeting, praying, among many others (PI section 23). His point is that speaking a language is part of an activity, of what he calls a life form. Here we cannot get away from the fact that the way we live life, and the language we use to describe life and act within it are not separate.

The words usually used to describe this intertwining of language and life is, perhaps unfortunately in English, ‘language game’. This might lend it a slightly frivolous air especially as Wittgenstein goes on to discuss the different sorts of activities that the word ‘game’ covers. But the point is a serious one: our different activities entail the use of different sub-sets of our language.

Consider, by way of an example, a zoologist taking their toddler child to a wildlife park and looking at a giraffe. The zoologist will see things differently from the toddler. The child will see a big animal with spots and a long neck. The zoologist will see adaptation to a savanna environment, and to eating leave on trees out of reach of other herbivores. The giraffe is the same for both people, but the interpretation is different, because of their different life experiences.

The point is that while the two people might explain what they saw in different words, the differences arise from their different contexts, the different environments and experience that a toddler and a trained scientist have had. It does not mean that the truth is subjective, that they both have a different truth about giraffes. The truths they have are compatible, just not the same.

Our different language games are not, therefore, incompatible. A zoologist can understand their child’s description of the giraffe, and the child, if taken slowly and carefully through the new ideas they may be presented with, can understand the zoologist’s description. Both are being honest about what they have seen and describing it accurately. Naturally, there may be issues when the child describes a lion instead of a giraffe, or the zoologist turns out to be a creationist with a non-mainstream description of how a long neck came about. But in general language works because both sides are honest, and misunderstandings can be dealt with by mutual goodwill and explanation.

The point here is that language is not, as most of us understand, experience and use it, totally transparent. For normal purposes we do not consider how language works – we just use it. But when we stop and think about it, language gets more complex. It starts to shape the world around us because part of that world is a human construct. For example, if we start saying that a formalised homosexual relationship is marriage, and people object, what the argument is over is the meaning of marriage. Whatever else it might be, marriage is a human condition, a human construct. Its meanings are what we make of them, as individuals and as society.

 Language, therefore, is not as transparent as we might like to think. An awful lot of meaning is driven by the context in which the words are delivered. When we start to think about it like that, we might start to be surprised that we can make sense to each other at all.