July 07, 2004

Don't Dangle Your Participles in Public

That's the title of one of Richard Lederer's syndicated columns on the English language. Like Lederer's columns in general, it's supposed to be funny -- he's fond of puns and other kinds of word play -- but it's also supposed to be instructive. (Lederer had a long career as a private-school English teacher.)

This one is a collection of writing errors, a vein that Lederer often taps. The fall into error is a brand of humor that tends to have a nasty edge to it, a superior They Should Have Known Better tone when directed (as in this case) towards professionals, a sneering They Are Ignorant Fools tone when aimed at ordinary folks, in particular students, with their extraordinary word choices, erratic spellings, and uncertain grasp of facts. I guess we're supposed to be shamed into learning some important lesson from these bad examples, but it would be hard for even a compliant and well-disposed reader to do much with what Lederer provides in this column.

Worse still, Lederer is, whether we like it or not, a significant public face of scholarship in English grammar. This is one of the main ways the literate (but non-linguist) world gets to see what we do. Oi.

Background: I came across this particular Lederer column on page 19 of the March 2004 Funny Times (a Cleveland, Ohio, compendium of recent humorous cartoons and essays, with what now counts as a way-left tinge to its politics). I posted a version of this critique to the American Dialect Society mailing list on March 23, but didn't put it on Language Log because it was so long. But now we're entertaining some longer and pretty detailed log entries, so I'm offering it. (You've been warned; you could bail out now.)

As for students and their foibles, look, I'm a teacher, and have been for forty years, and of course I sit around with other veterans and tell amazing war stories, but these occasions are suffused by sadness, since our students' failures are our collective failures too.

But on to the participles. Or, I should say, "participles", since only five of Lederer's fifteen Horrible Examples actually involve participles (two present participles and three past participles). The rest have modifiers of other types -- six prepositional phrases, one infinitival, one relative clause, one reduced comparative -- plus one pronominal reference example that involves modifiers only because the pronoun is inside one, though this example does superficially resemble classic dangling-modifier cases. The real theme -- twelve of the fifteen examples -- turns out to be attachment ambiguities. Surprise!

So here's Problem 1: The article is actually about modifiers in general, not just participles. This is just inexcusably sloppy for an English teacher offering grammatical advice in public; people are confused enough as it is about grammatical terminology, and now Lederer's column throws "modifier" and "participle" around as if they were synonyms.

The headline is a disaster in this regard, and so is the framing text:

"The AP Press Guide to News Writing advises: "The language has many ways to trip you up, most deviously through a modifier that turns up in the wrong place. Don't let related ideas in a sentence drift apart. Modifiers should be close to the word they purport to modify." These statements culled from newspapers and magazines demonstrate what happens when a writer dangles his or her participles in public:"

Horrible Example 1 has a wonky infinitival modifier:

(1) The family lawyer will read the will tomorrow at the residence of Mr. Hannon, who died June 19 to accommodate his relatives.

and we don't get to an actual participle until example 6:

(6) The burglar was about 30 years old, white, 5' 10", with wavy hair weighing about 150 pounds.

-- an attachment ambiguity that probably results from simple comma omission; certainly, it could be fixed by treating "weighing about 150 lbs." as the fifth in a series of descriptors and setting it off by a comma:

(6') The burglar was about 30 years old, white, 5' 10', with wavy hair, weighing about 150 pounds.

All five of the descriptors apply to the burglar, but there's no way to keep all of them close to "the burglar" -- though if you insisted on keeping "weighing about 150 pounds" right next to "the burglar", you could do it:

(6"a) The burglar, weighing about 150 pounds, was about 30 years old, white, 5' 10", with wavy hair.

(6"b) Weighing about 150 pounds, the burglar was about 30 years old, white, 5' 10", with wavy hair.

These versions, however, highlight the burglar's avoirdupois in a way that the writer almost surely didn't want, and they lose the nice increasing-weight effect of (6'), which puts the longest, heaviest descriptor last. (It would also be possible to shorten "weighing about 150 pounds" to "about 150 pounds" and move it up in the descriptor list.)

Horrible Example 15 might well be a simple comma error too:

(15) An ethnically diverse crowd of about 50 gathered at the Falkirk Mansion in San Rafael yesterday for a speakout against hate crimes organized by the Marin County Human Rights Roundtable.

Here there are two dependents on "a speakout": the prepositional phrase "against hate crimes" and the participial "organized by the Marin County Human Rights Roundtable". Once again, we cannot possibly get both of these dependents up against "a speakout"; we could move the participial up --

(15') An ethnically diverse crowd of about 50 gathered at the Falkirk Mansion in San Rafael yesterday for a speakout organized by the Marin County Human Rights Rountable against hate crimes.

This is clunky, because of the long-before-short order, and it's also potentially ambiguous, with a "Rountable against hate crimes" reading that strikes me as about as likely -- not very -- as the "hate crimes organized by the MCHRR" reading of (15). A much simpler fix would be to set off "organized by the MCHRR" in (15) by a comma; a somewhat more complex fix would be to move it forward and set it off:

(15"a) An ethnically diverse crowd of about 50 gathered at the Falkirk Mansion in San Rafael yesterday for a speakout against hate crimes, organized by the Marin County Human Rights Roundtable.

(15"b) An ethnically diverse crowd of about 50 gathered at the Falkirk Mansion in San Rafael yesterday for a speakout, organized by the Marin County Human Rights Rountable, against hate crimes.

Similar remarks apply to:

(13) Hunting can also be dangerous, as in the case of pygmies hunting elephants armed only with spears.

Still another participle Horrible Example, 12, is almost surely the result of that very common punctuation error, failing to use a comma between conjuncts:

(12) We spent most of our time sitting on the back porch watching the cows playing Scrabble and reading.

(I assume that we're not entertaining a reading in which the back porch watches cows, only a reading in which the cows play Scrabble and read.)

This is spiffily fixed with a single comma:

(12') We spent most of our time sitting on the back porch watching the cows, playing Scrabble and reading.

On the other hand, moving "playing Scrabble and reading" up to the front, real close to the verb "spent", produces a sentence in which our attention is strangely divided:

(12") Playing Scrabble and reading, we spent most of our time sitting on the back porch watching the cows.

One more example, this time with a prepositional phrase instead of a participial phrase: Horrible Example 9.

(9) Residents will be given information on how to reduce the amount of garbage they generate in the form of lectures, printed literature and promotional items.

The problem is to convey that it's the information that comes in the form of lectures and so on, not that it's the garbage that comes that way, or that the garbage is generated in those forms. In this case, moving the "in the form of..." phrase up will work nicely, so long as (once again) it's set off by commas, or the even stronger dashes (otherwise we risk the unintended reading "items on how to reduce..."):

(9'a) Residents will be given information, in the form of lectures, printed literature and promotional items, on how to reduce the amount of garbage they generate.

(9'b) Residents will be given information -- in the form of lectures, printed literature and promotional items -- on how to reduce the amount of garbage they generate.

Or we could leave the "in the form of..." phrase where it is, but set it off:

(9"a) Residents will be given information on how to reduce the amount of garbage they generate, in the form of lectures, printed literature and promotional items.

(9"b) Residents will be given information on how to reduce the amount of garbage they generate -- in the form of lectures, printed literature and promotional items.

Yes, the versions in (9") might still be misparsed, but they could probably succeed just fine in context.

On to Problem 2: As is so often the case, the advisors don't actually come right out and explain why there's an issue with the examples. Lederer and the _AP Guide_ tell you to keep your modifiers close to the things they modify, but they fail to mention why you should do that -- or, in fact, why the fifteen examples are (presented as) ludicrous. The missing word is *ambiguity*. Every single one of the Horrible Examples runs awry, at least in Lederer's judgment, because of an alternative interpretation that the writer didn't intend but the reader might fix on.

Example (1) can be read as saying that Mr. Hannon died to accommodate his relatives (which might or might not be true, but it probably wasn't what the writer had in mind), while the intended reading is that the reading of the will tomorrow is to accommodate Mr. Hannon's relatives. Fixing this can't be managed just by moving "to accommodate his relatives" closer to "read" than to "died", since other adjustments are needed; this is left as an exercise for the reader.

Example (3) could be read as describing a passer-by with a bullet in his head, instead of a bullet-riddled body:

(3) The body was found in an alley by a passer-by with a bullet in his head.

Once again, just moving the prepositional phrase "with a bullet in his head" forward (and setting it off with commas) is not necessarily the best move:

(3'a) With a bullet in his head, the body was found in an alley by a passer-by.

(3'b) The body, with a bullet in his head, was found in an alley by a passer-by.

The problem is the highlighting of the bullet in (3'). What really works here is the simplest of adjustments -- making the shot passer-by reading more unlikely by referring to the body as an inanimate object, via the pronoun "it", with no movement of setting-off at all:

(3") The body was found in an alley by a passer-by with a bullet in its head.

An equally easy fix is available for the following example, which (at least out of context) induces giggles:

(14) Police searched into the night for a man armed with a shotgun that walked into a Boulder pharmacy Thursday morning, demanded drugs and then fled.

This is another one with two dependents of a single head, in this case "a man". Moving "armed with a shotgun" after the relative clause doesn't help things at all, but a simple change in relative pronoun, picking out a specifically human antecedent, does the trick:

(14') Police searched into the night for a man armed with a shotgun who walked into a Boulder pharmacy Thursday morning, demanded drugs and then fled.

Problem 3: The advisors talk as if any potential ambiguity were a problem that needed fixing. This way lies madness. Attachment ambiguities and pronominal reference ambiguities are just everywhere. Telling writers to avoid potential ambiguities is pretty much telling them to give up writing entirely.

Potential ambiguities aren't the issue; effective ambiguities, ambiguities that a reasonable (and well-intentioned) reader might stumble on, are the issue. Picking out potential ambiguities requires (merely!) that you know the syntax of the language. Picking out effective ambiguities, which you might seriously want to avoid, requires factoring in common knowledge, your audience's mind-set, prosodic effects, the discourse context, and much else. Usage manuals are almost all piss-poor at this sort of thing. Lord knows what students are supposed to make of their advice.

Now, some of Lederer's examples really are laughable, at least out of context, because the unintended reading just juts out. The effect is especially strong for very short sentence-final modifiers, which tend to latch onto preceding stuff:

(5) Organ donations from the living reached a record high last year, outnumbering donors who are dead for the first time.

("...dead donors for the first time" would fix this, without movement, though movement works too. Setting "for the first time " off by a comma might work too.)

(7) Beginning with three games on Tuesday, the unmistakable drama of postseason baseball will grip all of us who love the game for a month.

("...all of us baseball lovers for a month" would do, and maybe setting off "for a month" with a comma. All of the movement possibilities for "for a month" sound awkward to me.)

(11) The dog was hungry and made the mistake of nipping a 2-year-old that was trying to force feed it in his ear.

(Moving "in the ear" forward pushes it between a verb form, "nipping", and its object, which is very awkward. Moving it before "nipping" is completely impossible, since "nipping..." is the object of the preposition "of", and you just can't mess with a preposition and its object. Major recasting is best: "the mistake of nipping the ear of a 2-year old that was...")

Problem 4: Examples are judged out of context, when in fact discourse organization, information structure, common knowledge, and the like play incredibly important roles in the interpretation of sentences -- a fact you might never appreciate from presentations like Lederer's.

Here's the dangler-like pronominal-reference example from Lederer's Failed Fifteen:

(8) Despite its dismal record in human rights, the House of Representatives has granted most favored nation status to China.

In isolation, it would be easy to (mis)interpret (8) as attributing a dismal human-rights record to the House. But in a discourse that's about China, the intended interpretation would come through without any problem.

Similar remarks hold about some of the examples above, not to mention sentences you can collect any day -- "...are trying to stop spam in federal court" (news report on NPR's Morning Edition, 3/11/04), "I found out what we'd done in '92" (line from the tv show Law and Order; in the context, it's absolutely clear that it was the finding out that happened in 1992).

Problem 5: Movement is offered as the only solution (because mislocation is identified as the only problem), despite the fact that, as I pointed out above, movement is sometimes not even possible (preserving the intended meaning), often not especially useful, and sometimes not the best solution.

Look at Lederer's deer-hunting example:

(2) Mrs. Shirley Baxter, who went deer hunting with her husband, is very proud that she was able to shoot a fine buck as well as her husband.

Moving "as well as her husband" after "shoot" favors the unintended meaning. Moving it after "able", and setting it off with commas, is possible but really awkward. The easy solution is not to reduce the comparative quite so much:

(2') Mrs. Shirley Baxter, who went deer hunting with her husband, is very proud that she was able to shoot a fine buck as well as her husband can.

This is not to deny that sometimes movement is the way to go. The one absolutely classic sort of dangling-modifier example in Lederer's list really needs the modifier moved, from:

(10) Aided by a thousand eyes, the author explains how ants navigate and how they use dead reckoning.

to:

(10'a) The author explains how, aided by a thousand eyes, ants navigate and how they use dead reckoning.

(10'b) The author explains how ants, aided by a thousand eyes, navigate and how they use dead reckoning.

Problem 6: Not so crucial as the others, but the assumption that what modifiers modify is a single word (rather than some larger constituent) can present puzzles for the student. Look at Lederer's sixth example (yes, this is the only one left):

(6) The suspect was spotted in a vehicle matching the description of one which had been stolen from the Annabelle area by Sheriff's Office Sgt. Craig White.

The textbook doctrine here is that the problem is that "by Sheriff's Office Sgt. Craig White" needs to be close to the word it modifies, "spotted" (how, by the way, does the student learn what modifies what?). And, in fact, moving it right after "spotted" does the job:

(6') The suspect was spotted by Sheriff's Office Sgt. Craig White in a vehicle matching the description of one which had been stolen from the Annabelle area.

But the clever student will notice that in (6'), "in a vehicle..." isn't close to the word it modifies. (Clever professionals understand that the problem with (6) isn't really separation, but an inadvertent attachment ambiguity.) Why isn't that bad?

An even more clever student will wonder why "The suspect was spotted in a vehicle by Sheriff's Office Sgt. Craig White" doesn't need to be reworded to (the more marked version) "The suspect was spotted by Sheriff's Office Sgt. Craig White in a vehicle". The short answer is that the "by"-phrase modifies a whole VP and that putting it after the VP is locating it next to the thing it modifies. It can occur after the head of the expression it modifies, if that avoids an effective ambiguity and/or allows a long and heavy constituent (like "in a vehicle matching the description of one which had been stolen from the Annabelle area") to come last in the sentence, but that's not its usual location.

So... we can guffaw at these examples, ripped out of their contexts and paraded in front of us under the banner of "dangling participles" perpetrated by writers who should have known better. or we can wonder, sadly, what a reader who's hoping to learn something from a well-known English teacher can make of all this -- confusingly labeled examples, a rule that drops from the sky, no guidance about when the rule is important and when it's not, instructions for repair that are hard to follow and often lead to very odd results. If you didn't already know how to write well, could you learn anything from Lederer's column?

That's a rhetorical question.

zwicky at-sign csli period stanford period edu

Posted by Arnold Zwicky at July 7, 2004 10:27 PM