I’m reading B.F. Skinner’s famous book Verbal Behavior at the moment. I think it might be the only extant attempt to make a serious go of understanding language in nonsemantic terms (that is, without thinking that what is special about words is that they mean things). It’s useful to me to see how far one can take the idea because I have spent the past while looking at arguments for scepticism about meaning and have come to think they have some force.
Skinner proposes to explain language, that is a speaker’s verbal behaviour, in terms of the favoured explanatory primitives of behaviourist psychology (conditioning, stimulus and response, reinforcement and so on). It should be conceded at the outset that there is a very big gap to be filled between the sorts of circumstances where stimulus-response conditioning seems readily applicable and the complex outputs of verbal behaviour. For example, to use an example from another book I have on the go at the moment, what explains the first sentence of Graham Wheeler’s Brew Your Own British Real Ale: “It has been some years since I wrote the previous edition of ‘Brew your Own…’ “? What catches me off guard here is actually the suggestion that any given sentence might have causes, as if there was nothing the individual could do about it once the particular combination of causes and influences had aligned. I think I am currently unprepared for this level of commitment to determinism, but that is perhaps just woolly thinking on my own part. Perhaps I am already committed to this much determinism because I believe that the mind is ultimately just the body and that all bodily movements have purely physical efficient causes.
To be sure, I can imagine causes for Wheeler’s utterance, namely that it had been many years since he wrote the previous edition and he wanted to say something true, and he’d been asked to write an introduction to a new edition of his book, and so on. But there are other ways of satisfying those requirements. It’s hard to think of a sufficient cause for that very sentence that doesn’t at least implicitly make reference to Wheeler’s conscious deliberation. But my understanding of the behaviourist programme is that it aims to explain these things without any recourse to anything going on between the ears. That just seems breath-takingly ambitious. However, maybe when I have read some more of the book, I will begin to see how such an explanation could be provided.
So far, I’ve been learning about the basic categories of the explanatory framework: the mand and the tact. These neologisms label two ways in which a verbal response (a word or phrase or sentence) could be bound up with the nonlinguistic physical and social environment. A mand is a response to an environmental condition (specifically a deprivation or an aversive stimulus) that, as it were, specifies its reinforcement. Examples are (uses of) phrases like ‘shut the door’, ‘turn that racket off’, ‘Milk!’, and ‘pass the salt’. In each case there is some need that drives the speaker to emit the response (desire to stop a draft, desire for silence, desire for milk, and salt ). The response elicits a reinforcing counter action by the audience which over time strengthens the response (makes it increasingly likely that the speaker will respond in the same way to the deprivation or aversive stimulus in the future).
Hugh is very ‘manding. He frequently mands ‘milk!’ (although increasingly these days he mands ‘milk, please’ I’m not sure what category ‘please’ falls into although it seems to fit the principles of conditioning well. Evidently, the use of ‘please’ is reinforced because Hugh’s parents respond more promptly and and favourably when he appends it to his mands. We also explicitly praise him (good boy for saying ‘please’) which we hope is reinforcing too.
The tact is a more conventionally semantic notion. A tact is a response which has a conventional correspondence with a discriminable stimulus in the environment and the response is reinforced when the correspondence obtains. Uttering ‘fire’ in the conspicuous presence of a fire, or pointing at the cup of tea and saying ‘hot’ or naming animals ‘dogs’ ‘cats’ etc, are all examples of tacting. Verbal behaviour is social behaviour and the way this operant seems to work is that different communities conventionally arrive at different responses that will be reinforced in the presence of the relevant stimuli.
This is very much under described, but I haven’t got very far with it yet. One thing I would say, from my observation of Hugh’s entry into language, is that tacting is reinforced in the presence of representations of objects just as surely as the objects themselves. For example, we agree and smile at Hugh when he says ‘dinosaur’, even though no dinosaurs really are present. Ditto for dragons. In fact, it is quite startling how unimportant the real existence of something has proved to be from a learning perspective. Even for things which really have existing counterparts now, like dogs, Hugh has learned tacts for things he has never seen. He’s seen quite a few animals actually thanks to zoos in Auckland, Dublin and Norfolk, but he’s never seen a Panda in the flesh, and can tact it quite successfully all the same.
Hugh tacts loads. One interesting series of tacts is his naming of the parts of our family. He indicates Dot and Ken and Frank and says ‘mummy’. ‘daddy’. ‘baby’ or ‘frank’ and then points to himself and says ‘boy’. I wonder if this will later form the basis for some kind of sense of the natural hierarchy of things.
What I like about the behaviourist position is the idea that things got to be the way they are by an evolutionary process of selection of behaviour on the basis of its consequences. This idea is pure gold. It has to be right and I’m confident it will be part of the eventual explanation.