Logic and Language

Some time ago I heard a grammarian say that grammar has nothing to do with logic. That is clearly an overstatement. Grammar is indeed not the same as logic, but it governs the use of language. Logic is a set of rules for assessing reasoning, and all sophisticated reasoning is carried out in language; so one set of rules is likely to have some impact on the other.

Most linguists, anxious to defend the double negative, would object to my saying that grammar governs the use of language, arguing that grammatical rules are just patterns abstracted from observations of usage; but the relation between usage and rules is more complicated. It works in two ways. Individuals learning a language spot patterns in the way others string words together, and apply those observed rules to the construction of their own sentences, so most of what is said is intended to follow a perceived rule, and when people believe they have identified a rule, they treat it as prescriptive, not just descriptive.

It is only because our linguistic behaviour follows rules that we are able to construct many sentences from relatively few words, and that is what makes our linguistic skills orders of magnitude greater than those of other animals Many other animals use signs, and quite a few can learn to recognise some human speech, but in their case the meaning of each sentence has to be learnt separately.

Rule following is a crucial part of linguistic usage, so that a particular usage that breaks a general rule is in a sense a departure from general usage however frequently people use it, so that it can make sense to say that even frequently used phrases are misuses of language.

It is possible for humans to learn individual sentences as the animals do, as in the phrase ‘splice the main brace’ which is normally used where there is no main brace to be spliced, but while a few quaint phrases of that sort have some charm, it would be most inconvenient to have a lot of them.

A case which struck me as odd even while at the primary school is what I’ll call ‘the disposable not’.

It irritated me that people would ask ‘Don’t you…?’ or ‘Won’t you..?’ when they meant ‘Do you ?’ or ‘Will you ?’ After all, if ‘Do you not?’ = ‘Do you?’, ‘You do not’ should = ‘You do’, making ‘not’ redundant. Similar observations apply to 'didn't' and 'isn't'.

I had no extra-linguistic motive, or at that age any philosophical motive, for questioning that usage. It just struck me as absurd. I was very puzzled that other people seemed too obtuse to notice the absurdity; I still am. In my youth I sometimes answered ‘redundant not’ questions as if people meant what they said - although I realised they didn’t.

Adult relative, simpering: ‘Isn’t Auntie thing kind to give you that sweet dear?’

I, thinking carefully of the meaning of the question, ‘No’

Although people were often surprised such replies they never saw the point. At the time I put it down to the obtuseness of adults. It still surprises me that people talk such a lot without reflecting on what they are saying or how they are saying it.

Inconsistencies in linguistic usage are like logical time bombs, posing a permanent threat to communication, because however well established some inconsistent usage may be, the standard meaning of the words in question can reappear at any time to disconcert us.

‘presently’ used to be related to ‘present’ as ‘quickly’ is still related to ‘quick’ (and if anyone asks whether “ ‘presently’ ” should begin with a capital P because it begins a sentence, the answer is no; it should begin with a capital quotation mark). Although very well established the contemporary meaning of ‘presently’ is fragile. The original meaning pops up whenever we read Shakespeare or talk to a student of English as a foreign language.

English is the international language of air traffic control, but it would be a rash traffic controller who used the word ‘presently’ to convey to the pilot of an approaching plane that it will be safe for him to land soon, but not just yet.