1

It shouldn't be controversial to say that the grammar of certain languages is more complex than what communication calls for. For instance, some languages have gendered nouns, and it is often unclear what is the communicative purpose.

Is there any consensus on why grammar becomes "more complex than needed"? If not, what are the leading theories?

7
  • Not so much a matter of grammar as of lexical distinctions, but Ken Hale once told me, in a discussion of subsection kinship systems in Australian languages (which are complicated far beyond any normal human need), that he thought what happened was that when somebody really smart and abstractly-inclined was born in an Australian culture -- somebody who we'd send to MIT to get a math PhD -- that person got interested in the kinship system and complicated it for intellectual pleasure.
    – jlawler
    Commented May 28, 2021 at 15:46
  • 1
    One reason for which noun agreement is used is to free word order to allow communicate further nuances (like emphasis). For me the fixed word order of English is highly unnatural, for example. So it does have some purpose - even if it is not the primary (which I think is the redundancy that Draconis mentions in their answer) Commented May 28, 2021 at 22:26
  • @jlawler Very interesting idea. It is easy to understand that an individual would complicate things, but it takes many to accept a proposed language change. Why would others go along with the MIT-PhD to be? :)
    – J Li
    Commented May 29, 2021 at 22:42
  • @DenisNardin There are certainly examples in which grammar is useful. I’m just commenting on the general observation that grammar can be more complicated, at times, than needed. Are you hinting that you think all (or at least most) grammar complications are justified by necessity in communication?
    – J Li
    Commented May 29, 2021 at 22:45
  • @JLi Well I'm not a linguist (that's why I'm commenting and not answering) but the idea that grammatical gender is "more complex than needed" strikes me as peculiarly odd: as a native speaker of a language that has it I really miss it in English. Commented May 30, 2021 at 5:34

4 Answers 4

6

Redundancy.

Speech (or signing, though there's been less research done on this) is an inherently imperfect communication channel. Information gets lost or corrupted very easily, and there's just no way to avoid it—communicating in a crowded party, or shouting across a long distance, or holding a conversation while music plays in the background, is often necessary.

As a result, human language has a whole lot of redundancy built into every aspect. Yu cn prbbly undrstnd Englsh wth all wrd-intrnl vwls ersd, even though that destroys a whole lot of information. Or you can hold a whole conversation in whispers (removing voicing information), or have entire words drowned out by noise, and still understand everything. This isn't a bug, it's a feature.

For your specific example (noun gender agreement), you can think of it as redundantly specifying what category a noun falls into, by putting that marking on adjectives and verbs as well as nouns. Or you can think of it as redundantly specifying which words go together, by marking them explicitly instead of just relying on word order. Either way, it doesn't (usually) add information you wouldn't get otherwise—it just adds redundancy in case one source of that information fails.

4
  • Agreed. Redundancy is a feature, not a bug. When you consider all the levels, language is something above 90% redundant, which is why statistical CL/NLP works so well. It's low-hanging fruit, and all human languages use it one way or another.
    – jlawler
    Commented May 30, 2021 at 15:34
  • I certainly agree that redundancy is part of the story, and I do appreciate the very clear and helpful answer. I do have to say that this leads to the question: “does language contain as much redundancy as needed for communication purposes?” We can get a sense of that from how computers transmit information over the internet. There are also redundancies designed to avoid information loss, but from what I know, those look nothing like the redundancies we see in everyday communication.
    – J Li
    Commented May 30, 2021 at 19:28
  • Further, this doesn’t explain the large differences of the degree of complexity across languages. Mandarin Chinese, for instance, has very simple and loose grammar. Their people communicate just fine. German has more complex grammar. There is no clear reason to expect that Germans have more need for redundancies than Chinese, no?
    – J Li
    Commented May 30, 2021 at 19:30
  • @JLi But is German grammar more complex than that of Chinese? Sure, it might have more complicated inflection, but German doesn't have Chinese tones, which many people have quite a lot of difficulty with when learning the language. Where one language's grammar may seem to lack complexity in one aspect, it makes up in another, again, complexity is difficult to pinpoint here. Commented May 31, 2021 at 5:51
2

Short and unambiguous

Complexity makes a grammar hard to learn, but not hard to use. Speakers of a language, having already learned the language, are not really interested in reducing its complexity. It is also not a conscious choice to adhere to the language's rules, so they don't feel "restricted" by them.

What an language speaker wants from their language is the ability to get their point across accurately and fast. So a grammar has to reduce ambiguousness while still allowing to form short sentences. That is what they need for "communication purposes".

If we measure complexity as the number of rules a grammar has, we see, that:

  • We need rules to make sentences less ambiguous: Forcing the word order to be SPO or marking cases on nouns gives an unambiguous way to parse a sentence.

  • Adding more rules to a grammar can also make it less verbose. For example, to be being irregular in English allows for the shorter I'm, he's etc.

But making a language less ambiguous also entails forcing speakers to give some extra information:

For example, you can drop the subject in Japanese and it has to be understood from context, sometimes leading to misunderstandings. German and English force you to specify the subject to counter this possible ambiguousness. This, on the other hand, makes the language slightly more verbose.

Most grammar rules I can think of reduce either verbosity or ambiguousness.

Regarding gendered nouns: German forces its speakers to mark the gender of almost every person as soon they are mentioned by using gendered nouns. So German speakers, in return, think it is very strange to hear from "a teacher" and not know if they're male or female.

This may not sound so useful, but gendering nouns comes at a very low cost in German: it is mainly marked on articles and adjectives that are already inflected for case, and the inflections are very short and easy to pronounce.

The biggest benefit of gendered nouns is that it reduces the ambiguousness of demonstrative, relative and personal pronouns, because only nouns with the right gender could be meant.

In the sentence

Das Pferd ging in den Stall und aß eine Karotte. Sie war orange.

meaning

The horse went into the stable and ate a carrot. It was orange.

the it can only be the carrot, not the horse. Using the other pronoun "es" we could have made it the horse.

To summarize: Complexity is not really a problem of a language's speakers, and adding more complexity can reduce ambiguousness and verbosity, which bothers speakers the most.

0

The question appears to assume that the language only exists as a method of communicating factual descriptions and making demands between humans. Once you consider the full range of things that language is actually used for, you should be able to see that mere communicative efficiency can be subordinated to other purposes, such as obfuscation and artistic expression. Language is as complicated as is needed, and the linguistic needs of humans goes far beyond low-level informational exchanges. It is as complicated as is necessary for all of the purposes of humans.

2
  • There are certainly more purposes to language. However, while such purposes can explain the variety of language constructions, they don’t explain why we enforce grammar rules to everyone that are more complex than needed. For instance, while it would be nice to allow people to use gendered nouns to achieve whatever literary purpose they so choose, it is unclear why we force every user of German to adhere to a set of “correct usage of gender nouns”.
    – J Li
    Commented May 29, 2021 at 22:50
  • @JLi Simply imagine if everyone just spoke as they chose, following but their own set of rules, would it make it simpler, or more difficult to understand as everyone's grammar was completely different and we had to adjust each time we spoke to a different person, everything has a purpose, even if that purpose isn't always clear. Commented May 31, 2021 at 5:56
0

I see everyone answering regarding the neutral reasons why grammar gets complex and stays this way, but no one wrote about the more "political" aspect, so allow me:

Grammar (and by extent, Orthograph), was often normalized when only rich/powerful people knew how to write. Back then, while matching the oral language was still the main goal, written complexities also helped to prevent the "commoners" to access a higher level of education. This doesn't apply to the grammar you can hear orally of course, but there are languages (for instance French) where most grammar doesn't exist orally.

Besides, even when the oral language evolved and sometimes got rid of old grammar rules, they often stay at the written form, again for social reasons.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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