I'm specifically asking: in "chicken soup", is the noun adjunct "chicken"
considered an adjective as a word or as a phrase? If my dictionary says a
word is a noun, does it just mean the word can be used as a noun or is most
commonly used as a noun, and nothing more, and similarly for an adjective?
This usually won't matter, but in the case where a word has both a noun and
an adjective form with different pronunciations ("arithmetic") it does. In
"arithmetic operator", "arithmetic" can apparently take on two
pronunciations depending on whether "arithmetic" is used as a noun-word or
as an adjective-word, is that correct? (Usually, of course, it will just be
the adjective "a'rithmetic", but could you force "arith'metic" as a noun
adjunct?)
S.
Whoops, swapped the two cases here. The noun's "a'rithmetic", of course, and
the adjective's "arith'metic".
S.
> Are "noun" and "adjectives" word classes or phrase classes or both or
> neither? (Excuse me if I don't have the terminology right, I probably won't
> have.)
>
> I'm specifically asking: in "chicken soup", is the noun adjunct "chicken"
> considered an adjective as a word or as a phrase? If my dictionary says a
> word is a noun, does it just mean the word can be used as a noun or is most
> commonly used as a noun, and nothing more, and similarly for an adjective?
What you may be missing is that nearly any noun can be used
"attributively" to modify another noun. In English, as in other Germanic
languages, we can chain together two or three or more nouns together in
this way. When we say "horse race" we don't have to say "Now the word
'horse' has become an adjective" -- we can just say it's doing one of
the things that nouns do.
>
> This usually won't matter, but in the case where a word has both a noun and
> an adjective form with different pronunciations ("arithmetic") it does. In
> "arithmetic operator", "arithmetic" can apparently take on two
> pronunciations depending on whether "arithmetic" is used as a noun-word or
> as an adjective-word, is that correct? (Usually, of course, it will just be
> the adjective "a'rithmetic", but could you force "arith'metic" as a noun
> adjunct?)
I would say that "arithmetic book" is noun-noun, and the stress is on
"rith". But "arithmetic mean" is pronounced with the stress on "mean,"
and yes, I would guess that dictionaries would classify that use as an
adjective.
Hm, Merriam-Webster 11 doesn't give any pronunciation for "arithmetic
mean" or "arithmetic progression." I think they missed the boat. They
only give one pronunciation for "arithmetic" as a noun, and they don't
give a separate entry as adjective. I expected them to have handled this
in *some* way.
--
Best -- Donna Richoux
I would be very surprised to hear "a'rithmetic operator" and, if I did,
would expect it to mean something like a person whose occupation it was
to operate a'rithmetic.
What you may have exposed is that, while we do regularly use nouns
attributively as if they were adjectives, we retain in some corner of
our language minds a sense of the adjectiveness of the attributive
noun, and will use the distinct adjective form, if one exists, rather
than use a noun attributively. In this case, a distinct adjective form
(the one with the accent moved later in the word) does exist, so we use
it.
This adjectival sense causes me (and apparently only me) some
consternation when I see a phrase like "women astronauts" or "women
legislators", for "woman" seems to me
to have become the only adjective in English that inflects for number.
Part of the explanation may be that the word is actually an attributive
noun, but that only helps a little, because we do not normally inflect
nouns used attributively for number, either. (It's almost always
"child actors", not "children actors"; it's almost always "mouse
droppings" and not "mice droppings".
Gary Williams
The Quirk et al. Grammar also gives "gentlemen farmers" and "menservants" in
the category of "Plural in both first and last element". If you count
compounds where only the first element is plural, Quirk offers among others
"attorneys general", "courts martial", "passers-by", "mothers-in-law",
"men-of-war" (i.e. ships). These don't seem to be the same sort of
construction, though, since the second elements are either adjectives or
phrases and couldn't reasonably be pluralised.
Alan Jones
> The Quirk et al. Grammar also gives "gentlemen farmers" and "menservants" in
> the category of "Plural in both first and last element".
Good work. And Google shows each of these favored by about 3:1 over
the invariable counterpart. I'd have been with the "gentleman farmers"
minority, but, if I ever had occasion to use the word, would probably
pluralized both elements of "menservants".
So I guess I can give up my angst over "women accountants", which is a
good thing, because it was distracting me from my angst over "one of
the only".
But why is it never "children actors"?
Gary Williams
Steven Pinker in _The Language Instinct_ mentioned that nouns used
attributively are much more likely to be pluralized if the plural is
irregular. So we have women doctors, teeth marks, men bashing, and flying
purple people eaters.
It is interesting that even though we write them as two words, the
attributive-noun-plus-noun form is usually a single word linguistically,
with a single main stress. Compare the pronunciations of "toy store" when
it means a shop that sells toys (in which the stress is on the first
syllable), and when it means a tiny store used as a toy (in which each
syllable gets its own stress).
In German the parallel construction is normally written as a single word
(e.g., Spielzeugladen -- "playthingshop"), which may be evidence of a
Teutonic flair for linguistics.