This may sound like a stupid query, but:

Is the indefinite article, a(n), a quantifier?

To this date, I've regarded it as a quantifier because, while it is an article, it quantifies an NP to indicate one of such exists. However, literature I referred to skips it and lists only 'two,' 'a few,' 'some,' 'a lot' etc. Please clarify for me.

  • 3
    Articles are typically called determiners; that's a class into which quantifiers also fall.
    – jlawler
    Jun 8, 2020 at 15:15
  • @jlawler Possessive pronouns are determiners, but I have my doubts they also qualify as quantifiers. Jun 8, 2020 at 16:14
  • There's lots of kinds of determiners, and they mutate back and forth. The indefinite article in most I-E languages that have it is derived from word for the number one, which is clearly a quantifier. Similarly, possessives and other determiners have very complex co-ocurrence rules; the makeup of the determiner constituent before the adjective constituent in an English noun phrase can get very messy.
    – jlawler
    Jun 8, 2020 at 19:40
  • 1
    @jlawler Hi, thanks. The definite article is not a quantifier, is it?
    – Sssamy
    Jun 8, 2020 at 21:54
  • 1
    Depends on the language. Not all languages have articles, and not all of those have definite ones (Turkish has only indefinite, for instance). In most I-E languages they come from demonstratives (Lat ille, illa), or PIE *to-. Again, if you want to call demonstratives quantifiers (I wouldn't), it's up to you.
    – jlawler
    Jun 8, 2020 at 22:14

2 Answers 2


The indefinite article surely is a quantifier -- as you say, it quantifies an NP to indicate existence and, more arguably, uniqueness.

The reason you haven't found it explicitly listed as such might be because in traditional formal semantics, "a" is often treated as synonymous to "some".


In many cases it isn't.

Arguable the question title is not counting anything

"is ... a quantifier"doesn't count anything.

The argument that there's an aspect of existence fails, because the aspect exists even without a determiner, eg. before uncountable nouns. If it is the default, it does not need mention. However, young children who have not learned object permanence would disagree, believing only in what they are seeing (peekaboo). So you might need a cognitive linguist to answer.

Another example, taken from German, fails in direct translation. Apparently the article is required: "I have a thirst". This sounds strange, because construction with I am is preferred. A strong preference for tangible objects with have can be observed. But it leads to the strange result that money is suddenly uncountable. The count just doesn't exist, for what it's worth, if it's none of your business.

That's still different from is a. If not a(n) meant "not even one", then the quantificantion is apparent. Indeed, wiktionary derives an from the same root as one. Also cp. no one, nobody.

However, translations of no one derive from deictic particles so frequently (in Greek, Kurdish, Hindu, German, neque) that this probably represents an original state of affairs, and it's a reasonable assumption that a(n) as well reflected a similar one, at least in part (cp. e.g. the derivation of it, or the inclusive particle in). Logicly you answer not "an", but "yes" (from one such determiner *ge-) to questions of existence, although you might answer with a precise wuantity. Although, it would be an etymologic fallacy to give too much weight to that. Vice versa, this question is interesting for the reconstruction.

All that aside, one could interpret the title as "part of the set", "of the type", for which the existence of the set is prerequisite, not introduced by the determiner.

I'd consider that distributive, because it can also be something else at the same time.

  • So your point is that because "a" can also act as something different, it can not be a quantifier at all? With this line of reasoning, it would be wrong to say that "frequent" is an adjective, because it can also occur as a verb. Jun 9, 2020 at 10:29
  • @lemontree as far as I know there's no concensus on participles being either or, but that's not what I'm getting at. I was going for a proof by contradiction about the fundamental essence of an a not to be quantitive. Of course you might argue it's three words in a trench coat … Insofar I'm leaning on the school grammar as it's mentioned in the question. Then I try to sketch out a whatsitcalled to let the etymology tank along with the school grammar category (what's it called: Wiktionary translates double mill or dilemma; this implies: trap; Guess I mean an ontologic dependency …)
    – vectory
    Jun 9, 2020 at 15:18
  • Of course downvoters will vote anyway about said dependency, although I'm aware that there's an etymologic fallacy to be avoided.
    – vectory
    Jun 9, 2020 at 15:19

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.