Arnold Zwicky's Blog

From a draft of a soliciting letter mailed to me recently, for my signature (details concealed):

[Institution X has made N grants to scholars over the years] – people who have and are making important contributions to [science].

(The problematic piece is boldfaced.) The draft was prepared by highly educated people who write a lot in their work, but still they came up with this example of a classic type of non-parallel coordination, with two conjoined complement-taking verbs (here, perfect have and progressive are) but a complement with a verb form appropriate only to the second, and not to the first (perfect have governs a past participle, progressive are a present participle: have made, are making). This is “government of verb form by the nearest”.

What I said to the colleague (and friend) who sent me the draft is that I have in fact studied the phenomenon, adding:

I don’t view it as a lapse in grammar, but a great many people do, and I would probably look foolish if this went out under my name.  Grammatical sticklers would insist on:

… people who have made and are making …

Actually, I used to view such examples as lapses in grammar, but over the years I have softened my assessment. As far as I know, I don’t use this sort of government myself, but I have come to think that for many people it’s not an inadvertent error, as many usage advisers have thought, but just an aspect of a grammar somewhat different from mine — a variant construction.

1. The usage literature. I’ll start by surveying some treatments of government of verb form by the nearest in the usage literature, then go on to look at some Language Log discussions of it and to the analysis of it.

The usage advice generally views the examples as involving illegitimate ellipsis (of the first complement verb), and hence of a failure to Include All Necessary Words. That is, the fault here is said to lie with the missing first complement verb; the literature “blames it on a word” (see here) and advises that this word be included.

The phenomenon comes up five times in the files of the Stanford OI! project. In chronological order:

(Fowler, H.W. 1926. A Dictionary of Modern English Usage. London: Oxford University Press.) From pp. 134-6: Verbs: do not omit second part of compound verb if its form changes: “No State has [adopted] or can adopt”

(Pink, M. Alderton. 1948. A Dictionary of Correct English. London: Pitman.) From pp. 122-3, on maintaining parallel structure: “always has (done), and will continue to do“: change of tense –> do can’t apply to both

(Bernstein, Theodore. 1958. Watch Your Language. New York: Macmillan.) P. 146: the understood word must be in the same form – number, tense, person – as the explicit one; insert other form in … “all men who have [served] or will serve”

(Bernstein, Theodore. 1977. Dos, Don’ts, and Maybes of English Usage. New York: Times Books.) From pp. 70-1: Ellipsis Slips: when the verb form changes, include both forms: “I have not [advocated], nor do I now advocate…”

(Copperud, Roy H. 1980. American Usage and Style: The Consensus. New York: Van Nostrand Reinhold.) Summarizing material from pp, 122-3: can’t omit the second part of a compound verb when there’s a change in number or form: filed can’t be omitted from “a firm has [filed] or is about to file for a franchise”

Then in a 2005 Language Log posting, I looked at two more recent handbooks:

(Lunsford, Andrea & Robert Connors. 1999. The New St. Martin’s Handbook.  4th ed. Boston: Bedford/St. Martin’s.) P. 263 mentions: “I had never before and would never see such a sight.”

(Hacker, Diana. 2004. Rules for Writers. 5th ed. Boston: Bedford/St. Martin’s.) P. 93 mentions: “Mayor Davis never has and never will accept a bribe.”

Both handbooks identify the problem in determination by the nearest as omission of words: “Add words needed to complete compound structures” (Hacker); “including all necessary words” (Lunsford & Connors).

2. Language Log. In addition to the long 2005 posting on parallelism in coordination just cited, I looked specifically at government of verb form by the nearest in one section of a 2004 posting, starting with “If you have or would apply …” reported to me by a colleague in e-mail (my colleague judged it to be straightforwardly ungrammatical) and going on to note further examples:

On ADS-L on 4/11/04, I reported that a grad student had sent me the example (from a friend in e-mail) “I could (and have) watched people play that game for hours.” Here there are two coordinated auxiliary verbs, requiring two different verb forms (base and past participle, respectively), which should (according to some hypotheses about reduced coordination) result in an irresolvable conflict, and ungrammaticality — but, instead, the form required by the nearer conjunct (the second one) determines the verb form. Similar examples can be found in more elevated contexts:

(Ex 1) There are plenty of venues at which Mr. Chirac could, and has, demonstrated his rapport with Mr. Schroeder. (New York Times editorial, “Playing Politics with D-Day”, 1/19/04, p. A20)

(Ex 2) To the Editor: The United States government’s attempts to manipulate the world price of crude oil by increasing gasoline taxes or by releasing oil from the Strategic Petroleum Reserve have and always will fail. (Richard J. Stegemeier, “retired chairman and chief executive of Unocal Corporation”, New York Times letter, 5/27/04, p. A30)

As I reported on ADS-L: “The grad student suggested I do a Google search on “could +and have” — which yielded over 8k hits (without going into Google groups)! Some of these are irrelevant, of course, but, still, the number is huge. Actually, “would +and have” netted over 17k hits. Even “might +and have” got about 2k. So there’s a hell of a lot of determination by the nearest going on in unguarded writing.

… I’d imagine that government by the nearest is also common in speech, given how incredibly frequent it is in writing. I’m tempted to suggest that government by the nearest conjunct is in fact the rule for vernacular English — which would explain why it’s so hard to teach people to avoid this construction in formal writing.”

Since then I’ve collected quite a few cases on the fly, including many in speech (in particular, from news reports on National Public Radio), many in informal writing (on mailing lists, blogs, newsgroups, and the like), and some in mre careful writing. A few examples:

Malibu tends to astonish and disappoint those who have never before seen it…  I had not before 1971 and will probably not again live in a place with a Chevrolet named after it. (Joan Didion, quoted by Jon Winokur in a column in Funny Times)

Analysts say that’s evidence Microsoft should – and likely is – taking Google much more seriously. (AP story in Palo Alto Daily News, “Google, Microsoft battle it out”, 11/22/04, p. 18)

…I bring Adrienne up to date on the high-speed moves in the direction of the nursing home and on the rational family conferences that should, but decidedly are not, taking place over this momentous decision… (Judith Levine, Do You Remember Me? (NY: Free Press, 2004), p. 222)

It’s a crime he says he did not and could not have committed. (Lynn Neary, NPR’s Morning Edition Sunday, 4/29/07)

That last example inspired me to do a Google search on “did not and could not have” on 4/29/07, which pulled up ca. 19k raw hits, many of them from administrative, legal, etc. contexts. As in this court opinion:

… to contest the medical necessity of a procedure after the fact, Trigon did not and could not have suffered such prejudice under the facts of this case. (link)

In combination with the difficulty of teaching people to avoid such examples in formal writing — students are often puzzled to be told that they have to replace “could (and have) watched people”, which they judge to be entirely natural, by the longer and clunkier “could watch (and have watched) people” — such frequent instances brought me to think that these elliptical coordinations represented just another variant construction in English, one that I don’t use myself.

3. Analysis. As I pointed out in my 2005 posting on parallelism, the handbook treatments of what I called there “factorable coordination” are inexplicit and confused, and I tried to clarify things some. In factorable coordination, involving expressions of the form X [Y and Z] or [Y and Z] X, conjuncts Y and Z (I’ll restrict myself to two conjuncts in this discussion) are associated with a factor X. One condition usually placed on (some types of) factorable coordination is

Distributivity: the factor must be syntactically well-formed in combination with each of the conjuncts.

That is, the syntactic well-formedness of [Y and Z] X depends on the syntactic well-formedness of both Y X and Z X. (There are several ways to incorporate this condition in a system of syntactic description, but for my purposes here, we can just treat it as a stipulation.) So government of verb form by the nearest is problematic, because in these cases Y X is not syntactically well-formed (though Z X is).

Another way of looking at things is that in government of verb form by the nearest there’s a conflict between the two sub-conditions of Distributivity, one of which requires the factor X’ that is syntactically compatible with Y, while the other requires the factor X” that is syntactically compatible with Z. The handbook accounts then assume Distributivity in the form above (in which it’s a condition on syntactic form), and also assume that conflicts between conditions are irresolvable and so yield ungrammaticality.

Occasionally, people maintain to me that these hypotheses not only hold for English, but necessarily hold for English: they are just a matter of “logic”. But there’s a considerable literature on types of coordination that run counter to one or both of these hypotheses.

In the case at hand, in fact, there are two lines of analysis (not very different from one another) for these syntactically non-parallel coordinations. In the first, the two conjuncts are required only to be semantically compatible with the factor (this condition is needed in any case), and the verb form of the factor is stipulated to be determined by the second conjunct. In the second, a conflict between the two sub-conditions of Distributively in government of verb form is stipulated as being resolved in favor of the second. (There is a considerable literature on principled resolution of feature conflict.)

The difference between varieties where government by the nearest is disallowed and those where it’s available as an alternative to fully parallel coordinations is then just the difference between not having vs. having an alternative coordination construction with one of these stipulations.

This entry was posted on December 3, 2009 at 11:48 am and is filed under Coordination, Ellipsis, Government, Syntax, Usage advice. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.