Subject: Right back atcha, Mr. Hyphenator

In the February–March issue of Copyediting, Linda Lowenthal writes about one of the most common stumbling blocks for copyeditors: hyphenating compound modifiers. In that article, she quotes an argument made by Bill Walsh in a blog post on Blogslot.com against a principle I use to determine whether some compounds need to be hyphenated. More on that in a moment, but first let’s describe the problem.

Hyphens help a reader know what parts of a compound modifier belong together as a unit and what parts are modifying the noun separately. An example I often use is white-veiled woman: with a hyphen, the woman is wearing a white veil, but we don’t know her race; without a hyphen, the veiled woman is white and we don’t know the color of the veil.

In the above example, the compound modifier is made up of two adjectives, each of which might reasonably be taken to be modifying the noun separately. Because of the ambiguity, it helps to use a hyphen to specify that the first adjective is modifying the second one rather than modifying the noun.

In some cases, what modifies what is completely clear. That’s why we don’t hyphenate compounds with -ly adverbs or those with very in them: we know that in a very gloomy mood, very is modifying gloomy, and that in a closely guarded secret, closely is modifying guarded. Similarly, capitalized compound modifiers don’t need to be hyphenated because we can readily see what the unit is: a United Nations resolution is not a united resolution of nations, but rather a resolution passed by the United Nations. We don’t need a hyphen to tell us that United Nations is a single unit.

Compound modifiers come in lots of combinations: adjective + noun (dark horse, snowy owl); noun + noun (town hall, maple syrup); adjectival participle + noun (marching band, split infinitive, chopped liver); and many others. Combinations such as those I’ve just cited are called fixed compounds: they have at least one discrete meaning, occur frequently and always in a fixed form, and have their own entries with definitions in dictionaries. Far more combinations are temporary compounds: I can write about a seven-step process but you won’t find an entry for seven step in a dictionary.

I have no quibble with using hyphens to join the parts of a temporary compound to show that they should be interpreted as a unit; doing so clarifies where the compound begins and ends. But I’ve long argued that fixed compounds don’t usually require hyphens for them to be understood as modifiers. A hyphen is a tool to use when ambiguity is a real risk, but when it is not — when no one would interpret maple syrup bottle as meaning the syrup bottle was made of maple — then using one is unnecessary. Insisting on ground-squirrel burrow, combat-fatigue symptoms, fox-trot steps, frying-pan handle, and the like is overly fussy, to my mind.

In saying this, I differ from Walsh, who would have us hyphenate high school cafeteria, presumably because we might otherwise think the school cafeteria was built on a hill, or on stilts. Linda Lowenthal says in her article, “Leaving compounds open if they’re in the dictionary is consistent — to someone who’s using the same dictionary as you, and who understands that this is why watercolor-painting classes is hyphenated but oil painting classes is not.” She goes on to thank Walsh for the example, citing his blog entry “How About It, Hyphen Haters?”

Here’s the full context, from a comment Walsh made on that post:

The permanent-compound approach (”law enforcement” isn’t in Webster’s New World, by the way) is flawed in at least a couple of big ways. “Oil painting” is in the dictionary, but “watercolor painting” isn’t, so you’d have “oil painting classes and watercolor-painting classes”? And some of those compounds are arguably onewordable — if rollercoaster seems like at least a possibly valid choice, I think it has to be “a roller-coaster ride,” not “a roller [dramatic pause] coaster ride.”

The idea behind the “onewordable” comment is that leaving a compound open makes its two parts seem less related, even though the open spelling is essentially arbitrary. (The fact that the two words have remained separate is simply an accident, judging by how many compounds have come to be spelled as closed over time.) Best to use a hyphen, Walsh argues, because it helps show how closely connected the words are.

But I don’t buy the “dramatic pause” argument. If the gap is so obtrusive, then why not insist that all open compounds be hyphenated or just closed up in all instances? Why insert a hyphen only when the compound is a modifier — and then only when it is in an attributive position, not when it comes after a linking verb? I think we come right back to the two core points, that hyphens are needed when ambiguity is possible, and not needed when it isn’t; and that they are needed to show where a temporary compound begins and ends, but usually not needed for fixed compounds.

I say “usually,” because in fact I would hyphenate oil-painting class. In the example of oil painting, Walsh is perhaps thinking, not without justification, about those who follow rules slavishly and treat dictionaries as holy writ — who might indeed determine that, since Wendalyn’s Principle is not to hyphenate compounds that have dictionary entries and to hyphenate ones that don’t, the right choice would be to hyphenate watercolor-painting class and not oil painting class. But dictionaries include or exclude entries for a number of reasons, and so the fact that a given compound has an entry can mean that one should consider the semantic pattern this entry represents in one’s determinations, whether all the possible iterations of that pattern are presented in a given dictionary or not. And in the particular case of oil painting, the primary meaning (an object, a work of art painted in oil) is so strong that it’s entirely possible that a reader would start to misread the modifier (which employs a secondary meaning) as a noun in its own right: “Free oil painting. Class. Oh, free oil-painting class.” Consideration of the context absolutely must come into play.

This means there isn’t a neat rule that will apply in all instances; as Lowenthal said, “Short of the hyphenate-everything approach, in fact, just about any system you devise risks looking inconsistent to someone.” If Walsh were to advocate the hyphenation of compounds whenever they are modifiers, regardless of position, that would be truly consistent. But I think even he would say that that would be overkill; and it would then seem arbitrary not to hyphenate all compounds all the time, and be done with it. All we would have done would be to shift the battle lines. Yet spelling is moving away from the hyphen: even the latest edition of The Shorter Oxford Dictionary made news a couple of years ago for its hyphen purge.

So the upshot is that Walsh and I draw the line in different places. I find the fixed-compound principle to be helpful (notice that I just hyphenated a temporary compound); I also, however, hyphenate compounds such as up-to-date when they occur after a linking verb, because they so often cause miscues on first reading when they are left open. I’m not worried about inconsistency — I don’t think many readers notice any system at all in the application of hyphens anyway, unless they’re used in ways that seem jarringly quaint. (As they do most of the time in fixed compounds, I would argue.) I’m more concerned about being unobtrusively helpful, allowing hyphens to do their job without calling attention to themselves.

If you enjoyed this post, make sure you subscribe to our RSS feed!