Re: Whence this form?

On Jul 28, 3:52 am, Eric Walker <em...@xxxxxxxxxxxx> wrote:
On Mon, 27 Jul 2009 21:07:12 -0700, Jerry Friedman wrote:
Where is it written that parts linked by "rather than" have to
match? /Garner's Modern American Usage/ (which I found on Amazon)
says on p. 670, "/Rather than/. This phrase can function either as a
/conjunction/ or as a /preposition/. As a conjunction (the more common
use), /rather than/ demands that the constructions on each side of it
be parallel: 'If we can, we will solve this problem diplomatically
/rather than/ forcibly.' But as a preposition, /rather than/ can
connect non-parallel constructions: '/Rather than/ staying home on a
Saturday night, we went out to six different bars.'"

So Garner, with whom I occasionally disagree (as I do here), says. But
Bernstein (_The Careful Writer_) notes this:

"If we start from the premise that in this phrase ['rather than']
'than' is a conjunction, modified by 'rather', an adverb, we must
conclude that grammatically the elements linked by the conjunction
should be parallel . . . . There is [a] . . . trend in usage,
however, toward treating 'rather than' as a prepositional group. . . .
Thus we get the idiomatic, though not entirely grammatical,
constructions exemplified in the following sentences . . . ."

So the choice to treat "rather than" prepositionally is, as Bernstein
puts it, "idiom over grammar".

"Grammar" has different meanings, and I'll try to go by the meaning an
author uses, but it's a very strange grammar that calls for a
construction that is, as Bernstein says, impossible:

*Rather than raised prices on all steel products, each company has
acted selectively.

Now that would all be very well did the idiom express something,
connotation or denotation, that the grammatical casting does not, or if
the grammatical casting were somehow manifestly infelicitous,

It is in certain cases, such as Bernstein's steel example above.

or if the
idiom had assumed the form called "cast-iron". But none of those is the
case. The classically grammatical form is clear, comprehensible to all,
[and] as concise as the other, [which is] an awfully long way away from being
universal practice.

Except when the matching verb in the "rather than" phrase would be
past or first-person singular present and would come either before the
main verb or more than about two words after it. Then the non-
matching forms are, as Bernstein says, "almost invariable". (I'm
surprised he needed the "almost".)

I repeat that--

"We can stop rather than continue."
"We can stop instead of continuing."

--are both sound, while--

Agreed.

"We can stop rather than continuing."
"We can stop instead of continue."

--are both unsound.

I disagree on the "rather than continuing" one.

Here's a sentence on doublespeak that he quotes (without commenting on
the grammar) on p. 268:

"It is language that conceals or prevents thought; rather than
extending thought, doublespeak limits it."

That is part of a long-ish (for the context) quotation on a completely
unrelated matter; I would have been surprised by any comment on its
grammar or style unless the thing was a true grotesquerie. In any event,
Garner has staked out his position: "rather than" can at times be
conceived as prepositional. It is perhaps noteworthy that Garner offers
no least comment on when or why one might see the phrase one way as
opposed to the other.

Neither does Bernstein, I gather--he just starts from a premise.
Follett offers no least comment on why he thinks "rather than"
requires matching parts. (Other phrases with "than" don't help, as
they have different syntax. For instance, you (always?) repeat the
subject in a sentence such as "The problem is that you need to plant
bulbs before you remove annuals," but you can't repeat the subject in
"She likes to plant bulbs rather than remove annuals." And there's
always the weird "than whom".)

The facts of usage show that this "rather than" construction is
anomalous and classical grammar fails to explain it. We can say,
"Doublespeak limits rather than extends thought" (though it's not to
my taste). "Extends" is the only possible form there. Classical
grammar would imply that we can say, "Doublespeak limits thought
rather than extends it" and *"Rather than extends thought, doublespeak
limits it." But that last isn't English.

So you have at least two options. You seem to pick the one of saying
that neither of the other two possibilities for that sentence is
right, so the sentence has to be recast. Another option is to note,
as Garner and Bernstein do, that the gerund is idiomatic there, and
use it if you want. You can infer from this, as they do, that "rather
than" is sometimes a preposition. And if it's sometimes a
preposition, maybe it can be a preposition other times too. After
all, many writers use it that way, including Garner. (This may be the
argument he doesn't give.)
....

I don't see any way to write that so the parts match.

"Rather than extend thought, doublespeak acts to limit it."

Okay, if you're willing to add two useless words of the kind that your
usage authorities are always deprecating (and I agree with them).

But as I said above, there's an anomaly. Usually it doesn't matter
whether a "rather than" phrase goes before or after the main verb.
This time it does.

More felicitous castings are doubtless possible, but that meets the need.

"Instead of extending..." might be better.

[skip to related follow-up posting]

Not in Garner, but I see Wilson Follett recommends "The lecture adds
much to the required reading rather than repeats it."

(As opposed to "The lecture adds much to the required reading rather than
repeating it."

That sounds terrible to me. . . .

Follett was noting a correction necessary to the thought as originally
cast, which is a different animal from recommending the correct form as
the best way of expressing the ideas in it. Something along the lines of
"The lecture adds much to rather than repeats the required reading,"
might do.

I agree that that that's much better than Follett's correction.

Or, if that is not yet suitably felicitous, other arrangements
can be made;

I'd sure make one. I don't like this construction at all when the
first verb phrase ends with a preposition or particle.

but at least it is grammatical and readily comprehensible.
(And try that with "repeating" if you want something that really sounds
terrible.)

I agree that "adds much to rather than repeating the required reading"
isn't great either.

I feel that, at bottom, the governing principle in choosing between
interpreting "rather than" conjunctively or prepositionally is Walker's
Law of Least Cervical Displacement. For those not yet acquainted with
it, it states that in choosing between alternative castings, each of
which is--at least in some quarters--defensible, the form that will cause
the fewest heads to snap up on reading it is the preferred form.

If, as I posit, the casting--

"Hopefully, we can do it during this generation rather than continue
to be distracted by the endless, well-funded parade of software tools."

--would cause no necks to be snapped, whereas the casting--

"Hopefully, we can do it during this generation rather than continuing
to be distracted by the endless, well-funded parade of software tools."

--might cause at least a few to be, the choice must be the first form.

I can guarantee you that the second would lead to some instances of
cervical displacement; I am pretty sure that the first never would.

I'm not so sure, depending on the audience. I'm happy with Walker's
Law, but I wonder whether it always gives the results you prefer.

--
Jerry Friedman
.

Relevant Pages

  • Re: "Could of" (Was: Be and Have in Arabic, English and German)
    ... preposition could follow a modal; a preposition could govern a VP; ... Several native speakers seem to have strong intuitions that in this phrase *is* indeed a form of a lexeme "of". ... The suggestion of a local reanalysis of just this one set of strings modal++Participle implies that this set of strings must have been represented in speaker's minds as a storage unit independently from other "have"+Verb collocations, although its composite meaning would have been computable from its smaller parts in a regular way, and thus its separate representation would have been redundant within the grammar. ...
    (sci.lang)
  • Re: Looking for a C program to parse CSV
    ... this is where the chips come down and we both confess that we have no actual clue as to how common or uncommon ending your sentence with a preposition is. ... We'd need to hunt down independent sources and get our facts straight, rather than making appeals to authority -- so we'll probably have to agree to disagree on the particular issue, and when I'm around you I'll have to be careful about what I end my sentences with. ... And what, pray, are any rules of grammar but "because thats how it ... What complicates the issue enormously, however, and what is at the heart of our disagreement, is the fact that language is man-made. ...
    (comp.lang.c)
  • Re: Cube Action?
    ... I do consider myself knowledgeable in English grammar, ... the OP's phrase "dropped after some consideration" is ... Also, your dictionary gives highly abbreviated definitions, and you ...
    (rec.games.backgammon)
  • Re: Swine flu
    ... English grammar. ... For your information, it is elementary English ... between is a preposition; ... even when there are multiple pronouns after the ...
    (soc.culture.baltics)
  • Re: Whence this form?
    ... "If we start from the premise that in this phrase ... puts it, "idiom over grammar". ... (As opposed to "The lecture adds much to the required reading rather than ... "The lecture adds much to rather than repeats the required reading," ...
    (alt.usage.english)
Loading