"Science" might not be the appropriate word but I think it conveys the gist of my question:

Is grammar always an agreed-upon, exact process among professionals when identifying usage and parts of speech (in the same way that you can rely on 2+2 to always equal 4), or can the lines become blurred when analyzing some of the more difficult aspects of it?
(For example, can there be a debate where a word's part of speech in a sentence can not be agreed upon, even though both sides have recognized and valid arguments?)

The reason I ask this is because I want to find better and more accurate ways of approaching and thinking about grammar. I am a beginner in grammar and I find myself thoroughly stumped attempting to analyze complicated and sometimes even simple everyday sentences. Most of the time, many of the words and usage of such words seem ambiguous.

I have always thought of grammar, and tried to approach it, as a black-and-white thing that can be solved with a formula. Is this correct? Is grammar black and white? Or do things get as muddy as they sometimes appear to me?

  • 3
    No. (And here are some more characters because SE has a floor on how short a comment can be.)
    – Dan Bron
    Apr 15 '15 at 14:20
  • 1
    @Dan en.m.wikipedia.org/wiki/Zero_width_space
    – bjb568
    Apr 15 '15 at 14:24
  • 4
    For what it's worth, even science is not an exact science these days. So if grammar experts don't agree about grammar, that doesn't necessarily make it a "soft" subject.
    – David Z
    Apr 15 '15 at 14:27
  • 1
    Particle physicists use the 5 sigma standard for the declaration of a discovery. So by implication if grammarians can discover rules of usage that are only breached on average once in every 3,500,000 utterances, one could say those rules reflect "scientific truth". But we know that language is mainly a haphazard outcome of time & chance, whereas many of us suppose there's something "organised" (by God, a few natural laws, or whatever) about what particle physics tells us. Apr 15 '15 at 14:37
  • 2
    This question would probably be better asked on Linguistics. But be warned, 'grammar' may be a bigger field of study than what you're thinking of. Apr 15 '15 at 14:40

I think that linguistics, the study of language, is a very scientific field, but the difficulty is in its massive breadth of data. Rather than being like classical mechanics it is more akin to the chaos of meteorology. Yes we can model the weather, but there's just too much data to produce specific narrow predictions with certainty.

Every speaker has a unique grammar of their language, which is called their idiolect. The idiolects of everyone who is part of the same speech community will be pretty similar, but there will be differences too. Adult speakers know tens of thousands of words, most with a large number of distinct senses, each sense with more and less prototypical uses, and the possibility to use varying kinds of idiomatic and metaphorical meanings. Your idiolect also controls how constituent parts are combined into sentences and larger structures. And that's just one person's idiolect! To write a grammar of the English language today you need to deal with the speech of billions of people, dozens of millions of speech communities, and the constant force of language change.

So often linguistics is fairly statistical. Sometimes a language will appear to have hard rules, other times they'll be much softer. For example, I don't think there's any disagreement about when to use him vs himself, and it's something that ESL learners seem to pick up pretty quickly too. But if you ask whether it's most natural to say "My mother and I went to the market." or "My mother and me went to the market." you'll get no consensus at all. This doesn't mean that language can't be studied scientifically, just that sometimes we have to admit we can't get a solid enough grasp on the data. But if you keep your scope of inquiry tight, or study minority languages, then the task is easier, and decent grammars can be written in only a few years.

  • 5
    +1 Good answer. But I would stress that terminology is probably far less consensual and defined than in meteorology: ask three linguists to define what a preposition is, and you will get four answers. One could argue that linguistics is more akin to the social sciences. Secondly, linguistics depends on the interpretation of language; it also requires skills and methods similar to those of the humanities, especially in some sub-fields. You often cannot properly analyse literature linguistically without understanding its complete and deeper meaning (cultural references, even in speech.) Apr 15 '15 at 15:42
  • 3
    Nice analogy with meteorology.
    – Mitch
    Apr 15 '15 at 17:44
  • @Cerberus Linguistics is definitely a social science. I think you're overestimating the lack of consensus about what a preposition is though. But it's definitely true for other terms. But if you're not dealing with language typology, and are just studying a single language, that doesn't have to be as big a problem. Apr 16 '15 at 0:29

If you're looking for a single unified approach, you might want to choose a different discipline.

Richard Nordquist: Grammar & Composition Expert, at
About.com About Education Grammar & Composition English Grammar

gives an overview titled

10 Types of Grammar: Different Ways of Analyzing the Structures and Functions of Language

He starts with

So you think you know grammar? All well and good, but which type of grammar do you know?

Quirk and Svartvik, in Investigating Linguistic Acceptability (1966), proposed a five-point scale for the degree of acceptability of individual grammatical constructions. But not everyone accepts this approach :-)

Not the answer you're looking for? Browse other questions tagged or ask your own question.