Conlang/Advanced/Grammar

This section covers advanced topics in grammar.

The universal grammar controversy

Way back at the beginning of the Intermediate level, we warned that the terminology we were about to introduce, analyzing the parts of language, was description of the form of language, not a recipe for building language — a map, not the territory. Now that we're up to advanced-level grammar, things get even more fraught. There is actually disagreement, often heated disagreement, between linguists who claim there are inviolable rules that human language must obey because those rules are hardwired into our brains as a built-in language facility — the universal grammar hypothesis — and linguists who... well, don't believe that.

The related descriptive idea of linguistic universals was developed in the late 1950s and early 1960s by linguist Joseph Greenberg. These are simply properties of human natlangs that appear to always, or at least usually, hold. A particularly famous one, which we touched on clear back in Beginner syntax, is that, of the three basic elements of a sentence, subject object and verb, in their default neutral order in any natlang the subject comes before the object. Although natlang exceptions do exist, it's unlikely you've heard of any of them unless you're a linguist — the exceptions that non-linguist English speakers might realistically have heard of are all conlangs (such as Klingon).

The universal grammar idea, that a built-in language-processing facility of the human brain determines how natlang grammar has to work, was advocated starting from about that same era by linguist Noam Chomsky, amongst others, and is sometimes known as Chomskyism. The heated controversy this idea can generate, on both sides, is also mixed with the question of whether language gives rise to meaning or meaning gives rise to language, with a side order of paleoarchaeology and how human language evolved in the first place. Adding spice to the mix, some linguists have blamed Chomsky's approach to grammar for a major shift of linguistics research away from field studies, thus depriving us forever of irreplaceable knowledge of natlangs that have vanished over the past half century; clearly that's not going to make Chomskyists and non-Chomskyists more friendly toward each other, either.

As a conlanger, you should keep two things in mind about these controversies:  (1) Know they're there, so you don't accidently step in them; and  (2) Good news! You don't have to take sides on any of it in order to conlang. There's no need for a conlang to adhere to principles of universal grammar even if its author is a devout believer in universal grammar. For one thing, a fictional language might be spoken by people who aren't human anyway, so that it wouldn't matter what constraints do or don't apply to human language; but regardless of fictional background, or lack thereof, there's a vigorous tradition in conlangs of experimenting with deliberate violation of accepted rules. It's not about whether you believe the rules have to hold; if you're convinced they do have to hold, it can still be valuable or just plain fascinating to find out what breaking them would entail.

What's in this section?

The parts of this section are:

• Aligning arguments: Morphosyntactic alignment — including theta roles, nominative-accusative, absolutive-ergative, split-ergative, Austronesian, active-stative, trigger, dechticaetiative.
• Forming words: Different strategies — including isolation/agglutination/inflection/fusion, polysynthesis, incorporation, oligosynthesis.
• Constraints on natural languages: Things that all natural-language grammars seem to have in common.
• Destroying the noun/verb distinction: Different ways to bend or break one of the most prominent constraints on natural languages.