"many a" -- singular or plural? Or just awkward?

[Arnold Zwicky]

On Nov 12, 2008, at 12:26 PM, Ann Burlingham wrote:

> ---------------------- Information from the mail header
> -----------------------
> Sender:       American Dialect Society <ADS-L at LISTSERV.UGA.EDU>
> Poster:       Ann Burlingham <ann at BURLINGHAMBOOKS.COM>
> Subject:      Re: "many a" -- singular or plural? Or just awkward?
> -------------------------------------------------------------------------------
>
> On Wed, Nov 12, 2008 at 2:35 PM, Arnold Zwicky <zwicky at stanford.edu>
> wrote:
>
>> a side point is the choice between "many a X" and "many Xs".  i
>> haven't looked at the matter at all, but i suspect that many people
>> see a subtle difference between the two.
>
> I do - I'm thinking that, say, "many a day" is less specific, more
> metaphorical? I would use "it's been many days since..." in a more
> specific manner, while "it's been many a day since..." would be less
> about actual days, and more an unspecified period of time.

i've been thinking about the topic since Ann B. wrote, and i talked
about it with Elizabeth Traugott at breakfast yesterday.  i think Ann
is on the right track here: there's a sense in which "many Ns" is more
specific than "many a N", so that this would be yet another case where
Y is X plus something (a topic on which i've posted on a number of
times; see a summary in
   AZ, 3/23/08: Y is X plus something:
  http://itre.cis.upenn.edu/~myl/languagelog/archives/005487.html ).

the trick is to say what the something else is, in some detail.
here's a beginning.

both elizabeth and i had the feeling that the difference was
aspectual, which caused me to think about Jan Rijkhoff's work on
"nominal aspect" ("Nominal Aspect", Journal of Semantics 8.4.291-309
(1991), and the book The Noun Phrase, OUP 1992/94).  the leading idea,
which has been around for some time, is that the telic/atelic
distinction in verbal Aksionsart is paralleled in several other
domains, in particular in the classification of nouns as "telic" count
vs. "atelic" mass and some distinctions in prepositions, e.g. "telic"
"to" vs. "atelic "towards".  (wikipedia has a pretty decent entry on
Telicity.)

Manfred Krifka characterizes the difference as involving "quantized
semantics" for the telics, "cumulative" semantics for the atelics.  in
the case of "many Ns" vs. "many a N", that would involve the Ns
treated as a quantum, a unitary collection ("many Ns") vs. just as a
accumulation of one N after another ("many a N").  on that view, "many
Ns" has the meaning of "many a N" ('more than one N') *plus* a meaning-
component that the individual Ns, taken together, form an aggregate.

(i haven't yet checked Rijkhoff's book to see if he takes up this
case.  this is still very much exploratory work, not a polished essay.)

some manifestations of this difference...  first, there are contexts
that require quantized semantics: subjects of "assemble", objects of
"count", for instance:

   Many linguists assembled in the ballroom.
     vs.: #Many a linguist assembled in the ballroom.

   I counted many linguists in the ballroom.
     vs.: #I counted many a linguist in the ballroom.

(the # marks semantic/pragmatic anomaly.)

this contrast is very sharp for me.  other contrasts are even sharper,
for instance in certain temporal constructions:

   Many days a month, I commute to Chicago.
     vs.: *Many a day a month, I commute to Chicago.

some contrasts are more subtle.  for objects of some other verbs,
"many Ns" is much better than "many a N" for me:

   I noticed many squirrels in the park.
     vs. ?I noticed many a squirrel in the park.

similarly in some expressions of spatial extent:

   Chicago is many miles from here.
     vs. ?Chicago is many a mile from here.

no doubt there are many more cases.  what i don't know is if there's
any literature on the subject.

arnold

------------------------------------------------------------
The American Dialect Society - http://www.americandialect.org