So, this past semester I've been working on a couple of projects concerning the pragmatics and syntax of Bengali and Oriya referents. I unwittingly opened up a can of syntactic and pragmatic worms in investigating referents using classifiers. Up until this point, I had wrongly and naïvely assuming that classifiers were more or less an East- and Southeast-Asian-only phenomonen (Bengali and Oriya border on Burmese, which also uses classifiers extensively, so I always assumed it was borrowed / substrate influence.) Apparently, although it's not common, it's not unheard of in other Indo-Iranian languages (Persian and Sinhalese also use classifiers, and they don't border on SE Asian languages). Wikipedia also notes Mayan, Athabaskan, Austronesian, and sign languages. Someone working in my group also noted that some African languages apparently use classifiers, too.

In Bengali and Oriya, classifiers are used after numbers, and their position in relation to the noun signal different levels of familiarity. For example, the classifier after the noun is similar to a definite article, and before the noun, similar to an indefinite article. In SE Asian and E Asian languages, they are used to mark individuation after words of quantity (like numbers and words like "many"), or are just obligatory for quantities. Hmong also uses classifiers for definiteness, too. Apparently, in some African language(s) classifiers can signal indefiniteness before the noun, too, according to my group.

My question is, then, how do classifiers work in languages that you're familiar with? Are they used with numbers and quantities, or can they be used by themselves? What meaning do they contribute to the noun, contrasting with a bare noun? Can they signal definiteness or indefiniteness? Does their positioning influence the interpretation of the noun in any way?

X-posted to linguaphiles and linguistics