Approaches to grammar [closed]
Rules of grammar don’t work like laws of Maths. Rules of Physics are slightly closer to rules of English ... in the way that jumping gets you closer to the moon.
English isn’t precise, and it isn’t consistent. You can make up rules that work in many situations, but you always have to watch out for exceptions. Here’s a classic example:
- time flies like an arrow
- planes fly like birds
Generalising, you might develop a rule that claims “X flies like Y” tells you how Y flies by comparing it to how X flies.
Then you come to:
- fruit flies like a banana
and the rule fails (because flies is part of the noun “fruit flies”, and “like” has the sense of “enjoy” rather than “comparison”).
So rules of English are merely descriptive of certain classes of words, phrases, and so on. In a way, this parallels the development of rules of Physics. People generated models from empirical observation. They didn’t always work, and when they did, they weren’t always precisely correct - compare Newtonian mechanics with models based on Relativity, for example.
The major difference is that English is significantly more malleable - just take a look at poetry. You can break rules of sentence structure, word order, etc, and still have a body of comprehensible text. So ‘rules’ of English will never be as rigorous as those of the mathematical sciences.
So-called rules of grammar differ from rules of mathematics or physics because they depend on human behavior. Rules of grammar are descriptive of how people actually use language. It may be more useful to think of descriptive patterns. Often patterns are described to help those learning a second language. The result may be entirely unfamiliar to native speakers, for example, the description of the order of adjectives before a noun.
This situation is different from, say, the identification of prime numbers. A number does not become more or less prime over time, but an English construction can become more or less common, or standard, over time; or its meaning may change.