(This is part 2 of a series of posts on the relationship between language and dance. Part 1, a very short introduction, is here. Part 3, which will look at units that languages are built of, will follow soon.)
Partnered dance is a liberating and thrilling thing to take part in. By requesting and acceding to a dance, the two of you commit yourselves to spending the next three minutes together, during which time you will each be in charge of one half of a four-legged monster, with the joint goal of making that monster move in a coordinated way, to music. When you start to achieve even a modicum of success, the feeling is truly exhilarating.
I think there’s a lot here that is similar to linguistic conversation. The superficial similarities are many. The interaction could take any number of different paths, none of which either of us will ever have trodden before. There may, of course, be particular things that one of us wants to say or do on that journey, but neither of us knows exactly how we’ll do those things (in dance, which moves to use; in language, which phrases). Also, just as there are two different roles in communication – a speaker and a listener – there are two roles in partnered dance – a lead and a follow. (The lead is typically a man, and the follow a woman, but that is only a convention.) However, exactly what ‘lead’ and ‘follow’ actually mean in this context is easily open to misunderstanding – and it’s here that the really interesting analogies start to show themselves. To see them, we first have to consider the different ways in communication is made possible.
Two models of communication
There are, broadly speaking, two ways in which communication can work. One is called the code model. Here, messages are encoded, transmitted, and then decoded. If the algorithms for encoding and decoding are appropriately calibrated to one another, then communication should be successful. The code model is nicely illustrated by the diagram below from Claude Shannon’s work A Mathematical Theory of Communication, although its origins go back far further.
The other way in which communication can work is called the inferential model. The work of a number of mid 20th-century philosophers of language showed that no version of the code model can capture the day-to-day realities of language and language use, where the same utterance can mean different things, and different utterances can mean the same thing, all depending on context. The inferential model was developed to describe such cases. In it, signallers do not encode meaning; instead, they provide evidence for it. The literal meaning of their words is part of that evidence, but so too are various other factors, such as tone, body language, and so on. These nuances give texture to our words, and without them we lose an important dimension of communication. Listeners must then take that evidence, and the context in which it is produced, and draw the best conclusion they can about the speaker’s intended meaning.
One thing that inferential communication makes possible, among others, is creativity in communication, and on both sides too. Signallers can choose to express their meaning with whatever evidence they wish to – so long as, of course, that evidence is sufficient for the receiver to interpret it in the way intended. For their part, receivers have choices about how to interpret that evidence. This is where humour comes from. Take the old joke that starts with one man telling his friend about his recent holiday with his wife. “Jamaica?” asks the friend. Now, the first man has understood this question, but chooses to interpret it in a different, creative way, and responds accordingly: “No, she came of her own accord”.
Is partnered dance like the code model, or the inferential model?
When people first think about the lead and follow relationship in terms of communication, they seem to think of something akin to the code model. The lead encodes the ‘meaning’ (something like: the instructions for what to do), and the follower decodes it. On this interpretation, becoming a good lead means learning the encoding algorithm throughly, and then using it without noise or error. Similarly, becoming a good follow means learning the decoding algorithm, and not making mistakes in interpretation.
But a better way to think about lead and follow is in terms of the inferential model. The lead provides evidence for what they have in mind for a section of music (different moves, or movements), and they might do this in a number of different ways (for example, a longer stretch, or a delayed movement). These differences can be of various sizes, and if they are sufficiently large then fluent users of the language would identify the two different forms as distinct things. Just as the same word can have two different meanings, even in the same context, if it is produced with two sufficiently different tones of voice, the same movement can communicate different things to a follow if it is produced in two different ways. For their part, followers must take the evidence the lead has provided, and choose how to interpret it. Certainly, there is scope for creative interpretation, just as there is in language. Some followers are rather literal in their following, but the very best play a more active role: there is almost always more than one way to interpret a lead, and the best following involves making those choices as best to fit the moment.
Contemporary pragmatic theory, which is based upon the inferential model of communication, emphasises how both speaker and listener play an active role in pursuit of a joint goal. In the case of language, that joint goal is comprehension; in the case of partnered dance, it is interpretation of the music. The role of music is of course a significant difference between language and dance, but underneath there is a more fundamental similarity, about how both parties play an active role in a joint activity. The communicative basis of each seems to me very similar.