Those pesky negatives (revisited)
Wilson Gray
hwgray at EARTHLINK.NET
Wed Aug 11 21:04:14 UTC 2004
On Aug 11, 2004, at 11:43 AM, Arnold M. Zwicky wrote:
> ---------------------- Information from the mail header
> -----------------------
> Sender: American Dialect Society <ADS-L at LISTSERV.UGA.EDU>
> Poster: "Arnold M. Zwicky" <zwicky at CSLI.STANFORD.EDU>
> Subject: Re: Those pesky negatives (revisited)
> -----------------------------------------------------------------------
> --------
>
> On Aug 10, 2004, at 12:01 PM, Wilson Gray wrote:
>
>> ...But, seriously, folks, I've been completely blind-sided by what
>> seems to me to be the out-of-nowhere emergence of the "splitting of an
>> infinitive" by inserting "not" that has now become common. I've racked
>> my brain and I can't recall that I was ever taught a prescriptive rule
>> against this. (FWIW, I've read only one prescriptive grammar in my
>> life and the only thing that I remember about it is that it was a
>> green hardcover, was written by a Jesuit, and was published by Loyola
>> University Press, Chicago.) There was no need for a rule against it
>> because NOBODY EVER DID IT! For all practical purposes, I never lived
>> anywhere but Saint Louis for the first quarter-century of my life.
>> Perhaps the non-occurrence of "to not VERB" was just a local
>> phenomenon or something. Oh, well.
>
> Some relevant observations:
>
> 1. The invented No Split Infinitives "rule" follows from a
> misapprehension about the syntax of infinitival "to", namely the idea
> that "to" combines with a following verb *word*, whereas in fact it
> combines with a following verb *phrase*. (I'm prepared to defend this
> analysis at some length, but i'll omit the defense here, trusting that
> ADS-L readers will appreciate the point.) Now, English VPs can contain
> initial adverbials --
> [You] [must [quickly finish dinner]].
> [Quickly finishing dinner] [would be a good idea].
> -- so that "split infinitives" like
> [You] [have to [quickly finish dinner]].
> are predicted straightforwardly from these two pieces of syntax: "to"
> combines with following VP, Adverbial can combine with following VP to
> make a VP. There would have to be some special constraint to *block*
> such "split infinitives", which in a sense come for free, given the
> rest of English syntax.
>
> And indeed felicitous examples are easy to find:
>
> (1) And the F.C.C. has abdicated enforcement of the “public interest”
> requirement in issuing licences. Time was, broadcasters had to
> regularly reapply and show public-interest programming to earn
> continuance; now... (William Safire, NYT, 5/22/03)
>
> (2) It was certainly good to finally see documents indicating that
> President Bush did not order the torture of prisoners. (NYT editorial,
> 6/24/04)
>
> (3) East Palo Alto’s City Council is set to backtrack on action to
> nearly double customer water bills after claims by a resident that
> state laws applied to the hike required more notice and a public
> hearing. (Palo Alto Daily News story, 6/28/04)
>
> Then, since VPs can begin with the adverbial "not" --
> [You] [must not [not go to church]]. 'You must avoid not going to
> church'
> [Not going to church] [would be a bad idea].
> -- we predict "split infinitives" with "not", like
> [I] [am going to [not pay attention to your complaints].
>
> For example:
>
> (4) "Are you going to not do something just because a *boy* doesn’t
> want you to?” asks Tiffany. (Michael Thomas Ford, That's Mr. Faggot to
> You (1999))
>
> 2. Speaking roughly, there are two ways to avoid such "split
> infinitives": preposing the adverbial from where it "belongs" to in
> front of "to", or putting it later in its VP ("to reapply regularly" in
> (1)). These alternatives are not always available: later positioning
> doesn't work in (2) or (3), or in any examples with "not", and pre-"to"
> positioning is at best awkward (for me, ungrammatical) in quasi-modals
> like "used to", "have to", and (as in (4)) "be going to".
>
> It's fairly easy to collect examples where people have preposed an
> adverbial to avoid "splitting the infinitive" and ended up with
> something that only a mother could love:
>
> (5) Mr. Blackburn added that the Panhandle Regional Narcotics
> Trafficking Task Force failed adequately to supervise the agent, Tom
> Coleman, in its eagerness to win battles in the war on drugs. (NYT news
> story, 3/11/04)
>
> (6) Listen to a radio personality who caters to a conservative
> audience and there’s a good chance you’ll hear an awkwardly unsplit
> infinitive. Paul Harvey, for instance, goes out of his way to avoid
> splits. I once heard him say something like “He went to Wal-Mart
> personally to thank the employees.” (Bill Walsh, The Elephants of
> Style)
>
> 3. In fact, many people discern a meaning difference between the
> preposed and the "split" variants, as in (7):
>
> (7) a. I want/expect not to be disturbed. [preposed]
> b. I want/expect to not be disturbed. [split]
>
> According to these people (most days of the week I'm inclined to be in
> their number), the preposed variant expresses an attitude (of desire or
> expectation) towards a state ('I want/expect it not to be the case that
> I am disturbed', 'I want/expect to avoid being disturbed'), while the
> split variant expresses an attitude towards an action, or a possible
> agent in that action ('I want/expect that I not be disturbed', 'I
> want/expect X not to disturb me'). The meaning difference is parallel
> to that between the two locations ("upper", postmodifying, and "lower",
> premodifying) for "not" with modals, both illustrated in "You must not
> not go to church" (above). (This is one of the ways in which
> infinitival "to" acts syntactically like a modal verb.)
>
> The meaning difference is subtle, and in many contexts the difference
> isn't important. But speakers and writers sometimes want to express
> the attitude-towards-action meaning, and in those cases the "split"
> variant will be just the thing for their purposes.
>
> 4. Coming Attractions: an argument that English does in fact have an
> intervention constraint applying to "to" + VP combinations (and to P +
> NP-object combinations and also to some V + NP-object combinations),
> though virtually none of the textbook "split infinitive" examples are
> of this sort.
>
> arnold (zwicky at csli.stanford.edu)
>
>
In general, I have no problem whatsoever with the splitting of
infinitives. I'm fairly certain that I was in high school before I even
heard about the no-splitting-of-infinitives "rule." Also, in general, I
try to avoid the splitting of infinitives in writing, but *not* because
I believe in the existence of some supposed "rule" against it. It's
merely a stylistic whim. However, this simply is not the case wrt "to
_not_ VP." For me, it is a genuine rule of grammar that "to not VP" is
wrong. I didn't learn it through being taught it or by reading it in
some prescriptive "grammar." I learned it from the fact that it was not
used by the people from whom I learned to speak.
I have to admit that this is not the first time that I've found myself
on the wrong side of the grammatical fence. I once believed that only
the illiterate or the semi-literate used "done" instead of "through" in
cases such as, "I'll be done with this posting in a few minutes" "you
can't go out till you're done with your chores" "are you done with
that?" "may I use that when you're done?" Then I realized that I was
wrong. The truth was that only illiterate or semi-literate *white*
people used such forms. Finally, I realized that I was still wrong, For
speakers of standard English, such use of "done" is perfectly
grammatical, used by speakers of all levels of education or social
standing or sophistication. Rather, it's the use of "through" in cases
such as those above that is a grammatical peculiarity, a feature of
everyone's favorite non-standard dialect, Black English.
Perhaps I should make it a point to remember a statement made by a
former classmate after she had challenged Morris Halle during a lecture
on a point of phonological theory: "I don't know why I said that! I
know that I speak with an accent!"
-Wilson Gray
More information about the Ads-l
mailing list