LINGUIST List 4.386

Tue 18 May 1993

Disc: Number, Pro-drop

Editor for this issue: <>

Directory

1. Paul T Kershaw, 4.364 Sum: Number marking
2. "Fernando Aviles", Pro-drop

Alexis_Manaster_RamerMTS.cc.Wayne.edu writes:

>
> In response to my suggestion that the following is a true universal:
>
> While there are languages in which plural is CONSISTENTLY
> more marked formally than singular, there are NO languages
> where the converse is the case.
>
> I have received no counterexamples.
>
I do not deny that this does, in fact, seem to be a universal. Even in those
languages where the plural is less marked for a certain (arbitrary or specific)
subclass of nouns (see my summary of a few weeks back), there are classes of
nouns where the converse holds true (i.e., where the singular is less marked)
or where neither is unmarked (logical "or" = v here). But I question the
utility of such a universal. Does it reflect a semantic reality which should
be reflected in the analysis, or does it merely reflect a cognitive tendency?
In other words, need a semanticist or a morphologist worry about such a
universal (especially due to its vague reference to consistency), or should the
concern be solely that of the socio- and antropological linguist? (This
question, granted, treads on philosophical thin ice, since it might
and has been argued that there is no semantic reality outside of cognition. Be
that as it may, however.)

-- Paul Kershaw Michigan State University KERSHAWPSTUDENT.MSU.EDU

 Dan Slobin offers useful criteria for those concerned with an
adequate definition of the unfortunately named `pro-drop'
phenomena, i.e. discourse function. He rightly observes that `much
more is involved...than simply setting the ``pro-drop parameter'',
which after all is but a structuralist formalism distinguishing
languages with person/number-agreement inflection from those
without, and, in itself, offers no explanation for why the setting
should be one way or another.
 His point that Spanish might better be treated as a `PRO-ADD'
language is well taken, since it is not, in fact, the case that the
subject is unexpressed in unmarked presuppositionally neutral
Spanish sentences, simply because no pronoun appears. On the
contrary, the subject is explicitly expressed in the inflectional
morpholgy on the verb, and, therefore, it makes little sense to say
it is unexpressed, or to speak of its having been `dropped',
unless, of course, English is held up as a standard of comparison.
 His observation that expressed pronouns are added and
syntactically positioned in Turkish according to discourse
functions such as contrast, emphasis, topic maintainence, switch
reference, and the like, shows that the setting of the paramater is
not simply a matter of `on' vs. `off' and suggests that these
functions should be taken into consideration in the definition of
`pro-drop'. It would probably be useful to proceed by taking the
unmarked presuppositionally neutral sentence as the most
appropriate context and proceed along the following lines:

Discourse Function Syntactic Marking Language Type
 Pro-Drop

-neutral response 0 [null pronoun] +
 to questions
 PRO [expressed pronoun] -

-offering 0 +
 information
 PRO -

 By consideration of the marked cases, the [+ Pro-Drop]
languages may be compared to determine the extent to which coding
strategies such as stress, reduction, cliticization, etc. are used
in lieu of, or as complements to, the syntactic, and morphological
expression of the discourse functions of topic switch, emphasis,
assertive contrast, and so forth. In short, cross-linguistic
comparison is best based upon features that are bound to be shared
by all languages, rather than upon language-specific morpho-
syntactic coding strategies, which are known to vary arbitrarily
from one language to another.

Robert Mix