16

For pretty much any grammatical category, I can think of a language in which it's a closed class. Japanese has closed classes of verbs and (verb-like) adjectives, for example, while Swahili has a closed class of (noun-like) adjectives.

However, I can't think of any instance of the class of nouns being closed. It makes intuitive sense for this to be a universal, since giving names to objects is a pretty fundamental part of human language use. But it's equally possible that I just haven't looked at enough languages to find a counterexample.

So: is there any natural, attested language in which nouns are a closed class?

P.S. By "closed class" I mean a class that resists adding new members, like English pronouns—compare the backlash against singular "they", which has been used since before Shakespeare, to the immediate acceptance of the new verb "to google".

Draconis
  • 65,972
  • 3
  • 141
  • 215
  • 2
    Some suggestions here: https://reddit.app.link/zgrztB8rUY – curiousdannii Aug 05 '19 at 16:29
  • It's easy to call a class with less than a dozen members "closed", but < 1000 members? -- I think something else is going on (a different kind of resistance). – amI Aug 05 '19 at 17:28
  • @curiousdannii Very interesting! Unfortunately I don't know enough about any of those languages to say if they actually have closed classes of nouns or not. If someone else knows more, it could make a good answer (even if it's "nouns are never quite as closed as other classes, but X language gets close, with feature Y"). – Draconis Aug 05 '19 at 17:37
  • 1
    How come verbs in Japanese form a closed class? If you can explain briefly... Very interesting question by the way! – lmc Aug 05 '19 at 18:30
  • 2
    @lmc I'm afraid I don't really know why, only that it is the case; new verbs tend to be expressed with nouns plus suru "do" (so one doesn't party, one goes to a party, or does a party, or whatever). Recently a very few new verbs have started to show up, though I'm not sure how widely-accepted they are… – Draconis Aug 05 '19 at 19:57
  • 1
    There are certainly cultures in which personal names are a closed class, however the recursive conjunction of parental names renders the idea moot. There are might cultures or sub-cultures which change so little, that no new nouns are needed. It would be more interesting to show that this cannot be, e.g. on grounds of universal grammar, in a mathematical way, or in language evolution theory. The empirical evidence over the open set of world languages cannot be reasonably exhaustive and would only serve as counter evidence, but hinging on the definition of "noun". – vectory Aug 05 '19 at 20:10
  • In theory it seems possible for a language to have a small, closed part of speech category 'noun', but which is modified by forms from some other open class in order to produce all the required entity-referring forms. I'm not aware of any such language, but this would parallel the many languages that have a small, closed class of verbs, modified by a large, open class (sometimes known as preverbs) to provide all the predications that are needed for human communication. – Gaston Ümlaut Jan 22 '20 at 21:22
  • @GastonÜmlaut That would make sense! If anyone knows of such a language, it would make a perfect answer. (For a pathological case, the constructed language Toki Pona does that, but it's not anywhere near naturalistic.) – Draconis Jan 22 '20 at 21:31
  • I guess it would be a bit like a noun classifier system but in reverse, with the classifying forms behaving in a way that allows them to be classed as nouns, and the specifying forms be assigned to a different POS. But (sadly) I don't think there is such a language. – Gaston Ümlaut Jan 24 '20 at 06:51
  • "Japanese has closed classes of verbs and (verb-like) adjectives" To google in Japanese is guguru, colloquial but quite widespread. – dainichi Feb 09 '21 at 05:17
  • Nouns are not a closed class. Definition: A closed class is a grammatical class of words with limited membership. These words have primarily grammatical meaning.

    Examples: (English)

    conjunctions demonstratives determiners pronouns auxiliary verbs https://glossary.sil.org/term/closed-class

     – Lambie Feb 24 '23 at 18:56
  • All languages invent new words all the time. Ergo, nouns cannot be a closed class. – Lambie Feb 24 '23 at 19:09
  • @Lambie That would seem intuitively true, but as mentioned in the question, e.g. Japanese has a (mostly) closed class of verbs, and uses new nouns to express new verbal meanings. Similarly, Swahili has a closed class of adjectives, and uses new nouns to express new adjectival meanings. It seems plausible enough that there's a language which has a closed class of nouns, and uses new verbs or adjectives to express new nominal meanings. Verbs and nouns (predicates and entities) are the two most basic classes, semantically; if verbs can be closed, why not nouns? – Draconis Feb 24 '23 at 19:19
  • @Draconis I do not know about every language on the planet but it seems very unlikely that nouns (if they exist in a language and they do in every language I have ever heard of) could be a closed class. And since languages have verbs and nouns, at least, no language can evolve if both those "classes" are closed. – Lambie Feb 25 '23 at 18:43
  • @Lambie Certainly, it would be a problem for both to be closed, but if a language can exist with verbs being closed (using nouns plus a neutral verb like "do" to construct new verbal meanings), I see no reason why a language couldn't exist with nouns being closed (using verbs plus a neutral noun like "action" to construct new nominal meanings). – Draconis Feb 25 '23 at 20:09
  • Because all I can find are that Japanese supposedly has closed verbs. And in English, in the literature, all closed classes have to do with function words or determiners, invariable words. Not with verbs, adverbs etc. – Lambie Feb 25 '23 at 20:13
  • @Lambie That's the case for English, but not for other languages. Swahili pretty unambiguously has a closed class of adjectives, for example. – Draconis Feb 25 '23 at 22:15
  • @Draconis OL, I will just say it. If you say that nouns are a closed class, you are saying new words cannot be added to the language. That has never existed historically. Words are added to languages all the time. – Lambie Feb 26 '23 at 17:38
  • @Lambie What I'm saying is, I don't see why. You could make that same argument for verbs and yet there are languages where verbs are a closed class. You can express new actions without introducing new verbs, by using a "light verb" with new nouns. It seems entirely plausible that there could be a language that does the opposite to express new entities without introducing new nouns. – Draconis Feb 26 '23 at 17:41
  • Languages don't "express entities". Language is referential and doubly articulated. If you can't refer to things of the mind or reality (referents) by adding them to your language as a word for a thing that previously was not there, then, you are not dealing with a (natural) language. And I'm done with this. :) – Lambie Feb 26 '23 at 18:04

1 Answers1

2

All languages which lack nouns would fall under that umbrella – noun is an empty class. A couple of example languages are Lushootseed and Riau Indonesian. To be sure, no language has a fixed class of expressions for referring to entities, but also no language has a fixed class of expressions for referring to actions. Appeal to semantic criteria is doomed to fail, because "part of speech" is not a semantic concept, it is a syntactic one. If the syntax of a language does not mandate a sub-division into "nouns" vs. others, then the class "noun" is so closed that it is meaningless.

user6726
  • 83,066
  • 4
  • 63
  • 181
  • "All languages which lack nouns would fall under that umbrella – noun is an empty class." Nouns are not usually called a class. – Lambie Feb 25 '23 at 20:17
  • Fascinating! I know very little about Lushootseed, but I believe it does have a (closed) category of pronouns? So you could definitely make an argument that it has a closed class of nominals, and uses other (not-specifically-nominal) classes to denote other entities. – Draconis Feb 26 '23 at 17:44
  • Yes, "pronouns are special" in the syntax. I only know about human pronouns, dunno if the poss-pro can be used in e.g. "its branch" referring to an inanimate. Maybe @jlawler knows. – user6726 Feb 26 '23 at 17:52