Hungarian

[MORAVCSIK Edit]

In connection with Bingful Lu's query - although, as Alan King pointed
out, not necessary in proper response to it - several features of
Hungarian have been identified in the discussion which indicate that this
language (as several others languages cited by various contributors) does
not always plural-mark words that have plural referents. This
message is  to give some other bits of evidence for Hungarian to make the
same point.

The relevant features mentioned so far are these:

   a/ Generic nouns:
      Jim McCawley pointed out that Hungarian nouns are generally in
the singular. This is true for store signs, as in Jim's examples, and
also for generic sentences such as "Apples are good for you." (in
Hungarian: "The apple is good for you."). - Actually, I am not sure that
this qualifies for "singular used for plural referent" since generic nouns
could be construed as having either singular or plural meanings. They can
be construed as semantically plural since they refer to all (or most)
tokens of a type but they can be looked at as semantically singular since
they refer to one genus. Thus, neither the use of singular, as in
Hungarian etc., nor the use of plural, as in, say, English, on generic
nouns can be said to be counter-semantic. Nonetheless, given Hungarian's
general preference for the singular as manifested in several other ways,
it may not be an accident that Hungarian opts for singular generics.

In my earlier message on the topic, I mentioned two other facts:

   b/ Paired body parts:
      singular nouns are used for a pair of body parts ("eye"
'two eyes'; "half eye" 'one eye')

   c/ Singular after numerals:
      The use of the singular AFTER NUMERALS ("two apple").

Here are some additional relevant points:

   d/ Verb agreement with numerated subject:
      After numerals, not only is the noun in the singular
but, if the phrase is a subject, so is the verb ("Two man is coming.")

   e/ Verb agreement with conjoined subject:
      After conjoined singular subject nouns, too, especially if human,
the verb is in the singular ("Eva and Mary is coming.")

   f/ Verbal prefix agreement:
      Those verbal prefixes that normally agree with the complement
of their verb in person and number may be either in the singular or in the
plural in reference to third person plural verb complements: (note that
the single or double dash following a vowel should really be over that
vowel)

   'He/she went after the boys.'

    EITHER:

      Uta'n -uk-ment a    fiu'-k -nak.
      after  P3 went the  boy  PL DAT

    OR:

      Uta'n -a -ment a   fiu'-k -nak.
      after  S3 went the boy  PL DAT

   g/ Possessum's number in possessive sentences:
      Semantically plural possessum-s may be in the singular in possessive
sentences, such as in "The boys have hat." meaning each has one.

   h/ Possessor's number in possessive phrases:

     l. If the plural possessor is a noun, as below,

        az  ember ha'z -ai
        the man   house S3

        az  ember-ek ha'z -ai
        the man   PL house S3

       *az  ember-ek ha'z -ai-k
        the man  -PL house S3 PL

        the possessum, which generally agrees with the possessor in person
and number, shows the singular for both singular and plural possessors.
It is like "the house-his of the persons" and not
"the house-theirs of the persons" as one would expect.

     2. If the possessor is a third person PRONOUN, the
plurality of the possessor is once again expressed only once in the phrase
but now it is expressed through the agreement marker on the possessum
and not on the possessor; i.e., 'their house' comes out as "his/her
house-their" instead of the expected "their house-their":

       az  o"     ha'z -uk
       the he/she house their

      *az  o"    -k   ha'z -uk
       the he/she PL  house their

On this, cf. also Turkish, Tatar, and Kabardian.

Thus, although Alan King is right in saying that the plural in
Hungarian is "generally obligatory" when a plural referent is meant,
there are several contexts where the singular is obligatorily or
optionally used instead. Apart from a/ and b/ above -
generics and body parts being in the singular -
these are all cases where the context of the constituent
that has a plural referent and is nonetheless singular-marked
includes some other indication of the referent being plural. This
indicator may be a numeral (c/), the subject (d/), the
complement of the verb (e/), the subject of the possessive
sentence (f/), the possessor (g/) or the possessum (h/). Thus,
the facts under c/-h/ are all cases of economy: the LACK OF
DOUBLE MARKING of plurality (even though double marking would
be expected by general rules of agreement) - rather than
cases of NO MARKING of the plural.

Edith Moravcsik