14

Given:

  • For those languages which have it, the indefinite article mostly if not always is derived from the numeral for "one".
  • Most languages have numbers but many lack articles.

How do linguists decide whether a language under investigation has an indefinite article as opposed to just having a word for "one"?

hippietrail
  • 14,687
  • 7
  • 61
  • 146
  • great question! but did you mean 'indefinite' in the final question? – James Tauber Oct 02 '11 at 16:59
  • 1
    at the other end of the spectrum I wonder what the diagnostics are for demonstratives versus definite articles (as the latter often comes from the former) – James Tauber Oct 02 '11 at 17:00
  • @James: Yes I did, sorry for making so many typos/thinkos in today's questions )-: An your opposite end of the spectrum question is also very good, I think you should ask it. – hippietrail Oct 02 '11 at 17:46
  • 1
    see http://linguistics.stackexchange.com/questions/654/what-diagnostics-distinguish-demonstratives-from-definite-articles – James Tauber Oct 02 '11 at 18:00

2 Answers2

5
  • An indefinite article forbids bare nominal arguments (like bare nouns) when the nominal arguments are indefinite.
    • For example, in English "*I saw bear" is not valid, while in Mandarin (there is a word for one but no indefinite article) "wǒ kànjiàn xióng le" (I see bear ASP) is valid.
    • In French (with a plural indefinite article "des"), *"Il y a livres là-bas" (lit. There are books over there) is not not allowed while in English (no plural indefinite articles), it is OK to say "There are books over there".
  • An indefinite article adds no semantic content. For example, when you say "I have one printer", you probably mean that you have exactly 1 printer, while in "I have a printer", a is just used as an indefiniteness marker without adding any semantic meaning.

Reference: Indefinite articles and beyond - BSW Le Bruyn

hippietrail
  • 14,687
  • 7
  • 61
  • 146
Louis Rhys
  • 8,501
  • 6
  • 45
  • 71
  • 1
    Your first point (as made by Le Bruyn) is incorrect, there are languages with indefinite articles that do not require their use (ie they allow bare nominal arguments). – Gaston Ümlaut Oct 03 '11 at 09:20
  • 2
    @GastonÜmlaut can you perhaps cite an example? Why is the "article" considered an indefinite article instead of just a determiner (like English "some")? Does the language have a definite article? – Louis Rhys Oct 03 '11 at 09:59
  • 1
    sure. Wutung (Sko family, Papua New Guinea) is analysed as having indefinite articles (sg and pl) but no definite article. It also has the category of determiner, but this is a higher-level category that groups together several sub-categories (the articles, quantifiers, demonstratives and numerals). – Gaston Ümlaut Oct 03 '11 at 13:01
  • 2
    It's unclear that the indefinite article "a" in English "adds no semantic content". Because it is possible to say "There was bear on the road", which could be used to describe a situation in which a bear was hit by a semi-truck and now "bear-stuff" is all over the road. This contrasts with "There was a bear on the road", where what is indicated is a singular instance of the kind bear. "I saw bear" should thus not be starred except on the interpretation "one bear", since "I saw bear at the market today" (shockingly! bear-meat) has a clear sense. – Alexis Wellwood Oct 03 '11 at 13:42
  • 1
    @AlexisWellwood I believe in your example, there are two different word ("bear" the uncountable noun and the countable noun). The "a" unintentionally helps resolve the ambiguity because we know that it (syntactically) can co-occur with a countable noun but not with an uncountable noun. So the role it is playing there is syntactic, not semantic. – Louis Rhys Oct 05 '11 at 12:57
  • 1
    @LouisRhys See the Encyclopedia of Philosophy entry "Nouns, mass and count". If you claim ambiguity between count and mass occurrences of nouns, you are (at least) in the uncomfortable position of stating why it is that it is crucially stuff/units falling under the concept BEAR in both cases, why it is not a bank/bank type ambiguity. And we can multiply examples, so this approach is not explanatory. Rather, syntactic context forces a certain perspective on a (unitary) concept: I saw a steak (at least one unit) vs I saw steak (uncommitted to # of units). – Alexis Wellwood Oct 05 '11 at 13:47
  • @GastonÜmlaut I know it's a very late reply, but just in case someone reads this there is a closer-to-English example in Swedish, where we allow constructions corresponding to "I saw bear". In sentences like "He has car", "I live in house" or "I run farm", where English speakers always include an indefinite article, Swedish speakers would normally drop them. It's not just an example of sloppy speech either: Including an indefinite article to these examples is possible in many contexts but it doesn't make them more formal, in some circumstances adding an article makes it seem less formal! – EdvinW Jul 16 '22 at 21:41
4

An indefinite article is a form with a certain set of functions. These functions are in a number of respects different from those of words for the number 1. Linguists look at samples of language, determine what forms exist, and determine what set of functions they fulfill. For convenience these forms will be referred to by a suitable name, based on the analysis.

Beyond this it gets a bit complicated, but here's some parts of it, focussing on properties of indefinite articles that distinguish them from the numeral 'one'. In many cases the analysis of indefinite articles will involve looking at distributional properties: where in the noun phrase (NP) does it occur, and what other elements can it co-occur with. Articles normally don't occur alone, and act as modifiers of a NP. Indefinite articles may indicate number (eg singular/plural). Semantically, articles indicate whether an item of information is a specific individual, or if it's known to the hearer or not.

hippietrail
  • 14,687
  • 7
  • 61
  • 146
Gaston Ümlaut
  • 6,697
  • 30
  • 46