3

English has a few determiners and determiner-like things that express universal quantification including all and every.

How naturalistic is it to have different sets of quantifiers based on features of the domain such as its number (and possibly other features like animacy or remoteness).


The noun governed by the determiner can be either singular or plural with little change in meaning.

Every rose has thorns.
All natural numbers are odd or even.

These sentences have the same meaning as All roses have thorns and Every natural number is odd or even.

In some languages, such as French, have something that can be analyzed as a determiner inflecting for gender and number of the noun: tous/toute/tous/toutes which means all. Although French also has chaque, analogous to the English every and each. The distribution is similar to English, but I can't provide any examples that I can independently verify because I'm not a native French speaker.

Anyway, I'm wondering about the naturalness of having different determiners depending on the number of the domain rather than the thing directly governed by the quantifier.

For instance, suppose we had every(0) every(1) every(2) every(paucal) every(plural)

every(2), every(paucal) and every(plural) are all(paucal) fairly natural.

every(2) seems to map onto English both more or less directly, although both cannot be used with a singular noun.

every(paucal) continent has a name starting with a vowel or a glide.
every(plural) electron is small.

However, a construction like all(0) does have uses that aren't jokes and aren't related to logic or math.

A construction like the following seems reasonable and fairly easy to interpret.

I have done all(0) of my homework.
1
  • 1
    I would argue that as you define it, every(1) is just a special case of definiteness, with "Every(1) sun is bright" meaning the same as "The sun is bright."
    – Andrew Ray
    May 6 at 17:31
4

This is not inherently unnatural. As you point out, English "both" (and reasonable translations such as Spanish "ambos") easily fit the role of every(2) (I would argue that whether the noun is marked as singular or plural isn't particularly interesting here).

every(1) can be reasonably analyzed as a definiteness marking and is encompassed in most Indo-European languages by the definite article ("the", "el", "i", το), as shown by the sentence "The sun is shining." (This sentence can be used without explicitly introducing the sun into the conversation specifically because there is only one reasonable referent.)

In particular, I could see one set of forms evolving from the definite articles in a language with heavy number marking (i.e. "the (singular)" becomes each(1), "the (dual)" becomes each(2), etc.).

What I would be very surprised to see in a natural language is a fully consistent system of these, where each(), every(), and all() all cover the same numeric distinctions in the vein of the Esperanto correlatives.

I would also be surprised to see more than a couple of these in any language that does not make at least as many number distinctions elsewhere.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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