Ah, the art of hyphenation. It is an art, not a science. Compulsive hyphenation is in deserved disrepute, but no two authorities agree on which compounds in their noun forms are closed (dataset), hyphenated (data-set), or open (data set), and even more contentious, which ones to hyphenate when used adjectivally. Words seem to evolve from open compounds through the ambivalent hyphenated version into closed compounds, but lately, hyphen-allergy has been driving the evolution both ways. Open compounds are not hyphenated when one has a gut feeling they should be. I don’t know how many articles I’ve worked on that say “in vitro studies” rather than “in-vitro studies” (note to those who think it should be italicized: once a word’s in Webster’s, it has joined the English language and achieved roman citizenship).
Sometimes such “open” adjectival stylings, even when they don’t create any ambiguity, feel ponderous and floppy to me. They slow a sentence down, burden it, make it blank-faced like someone on Haldol and less nimbly expressive. The hyphen, by hitching the two parts of the compound together, lightens it up and make it move faster. (This stuff is all visceral and kinesthetic to a copy editor. Or should that be copyeditor? The former sounds slower and more deliberate, as copy editors should be, even though we should also work quickly to keep costs down; so I’ll stick with it, except in my blog title, where . . . oh, never mind.)
That’s an example of my ambivalence about the hyphen. Looked at more closely, it’s not really ambivalence, it’s context sensitivity. We are told as copy editors to be consistent: either all hyphens in a given grammatical (e.g. adjectival) situation, or none. But too many hyphens looks cluttery and fussy and Victorian; too few gets blank-faced and sometimes leads to impermissible and hilarious ambiguity. What I find myself wanting to do is hyphenate, or not, on a case-by-case basis. (Here, “a case by case basis” would make me feel as if I was talking in slo-mo, a 45 played at 33, for those old enough to get the reference.) But then it gets too subjective.
Here’s an example—a passage from what I was copyediting tonight. (Unlike the noun, the verb is usually a closed compound, which enables us to be deliberate yet work fast.)
But beginning in the 1980s, peppered moth experts including Majerus began noting flaws in Kettlewell’s experimental designs. The most serious of these was that limited research into peppered moth behavior seemed to suggest that tree trunks were not the insect’s preferred resting place. That alone threatened to put a serious dent into the validity of Kettlewell’s setup, as well as the bird predation theory itself.
In his 1998 book, Melanism: Evolution in Action, Majerus discussed these shortcomings along with a critical dissection of all the peppered moth case evidence that had accumulated.
The compound at issue is “peppered moth,” used adjectivally. I have argued elsewhere on this blog (it’s late, I’ll find the link another day) that compound common names for animal and plant species become units in our minds and thus should be treated like “lowercase proper nouns.” For example, despite its admitted grammatical ambiguity, “bald cypress swamp” will not be taken for a bald swamp with cypresses in it, and should not be hyphenated simply because if you once start to hyphenate things like that there’s no end to it.
But in this case that didn’t feel right to me. Going by feel alone, here’s how I wanted to edit the passage, and the footnote presents my case for it:
But beginning in the 1980s, peppered-moth experts, including Majerus, began noting flaws in Kettlewell’s experimental designs. The most serious of these was that limited research into peppered moth behavior seemed to suggest that tree trunks were not the insect’s preferred resting place. That alone threatened to put a serious dent in the validity of Kettlewell’s setup—and in the bird predation theory itself.
In his 1998 book, Melanism: Evolution in Action, Majerus discussed these shortcomings in the context of a critical dissection of all the peppered-moth case evidence that had accumulated.
 (CE) My reasoning is to use a hyphen where there is grammatical ambiguity, which, even if it doesn’t confuse us as to meaning, makes us feel funny. There is no such thing as peppered behavior, so that doesn’t call for a hyphen, but there could conceivably be a peppered expert on moths [as well as a peppered moth-case, before you even get to the evidence]. That grammatical form exerts a pull strong enough to rival the habitual bond between “peppered” and “moth.” In my opinion. If you think it reads as inconsistent, and “peppered-moth behavior” should be hyphenated too, go ahead. The trouble with that is that one then feels a compulsion to hyphenate every occurrence, and it becomes a nuisance.
As I look at this I realize that, another layer deeper, whether a hyphen feels needed or superfluous has something to do with how the sentences move and sound, where the stresses fall: “peppered móth expert” seems to throw “moth” and “expert” together, which is perfectly appropriate in the case of “peppered móth behávior,” but in the case of the expert, leaves “peppered” too equidistant from the two animate nouns, attracted to them both. (So why does this not apply to “bald cypress swamp”? Do I contradict myself? Or is it that the accent on “cy” juts up so high, like a cypress knee, that it shoulders “bald” and “swamp” far enough apart? No need, to my ear, to give the words a hit of helium and make it “baldcypress,” as some do.)
I think this probably goes too far into editorial prima-donnaism. After all, let’s get real, copyediting isn’t Beethoven’s late quartets, even when you have been doing it that long. Really, the problem is that I’m drunk on sleep deprivation. Consider this a rare glimpse into the place deep in an editor’s mind where language meets music and dance.