John McIntyre on his Baltimore Sun blog:

MARCH 18, 2011

When the word went out today that the editors of The Associated Press Stylebook would announce changes in AP style at the national conference of the American Copy Editors Society, the nation ground to a halt.

Factories suspended production. Police and fire departments called in employees on overtime. Members of the Cabinet were summoned to the White House. Knots of anxious civilians gathered in the streets to speculate worriedly on the decisions about to be handed down. Some ducked into bars for fortification.

And then the tweets began scattering across the Twitterverse:

E-mail will become email as of 3 a.m. EDT on March 19, 2011.

The sensation that this announcement sparked cannot be described.

But there was more.

Cell phone and smart phone fuse into single words.

Gasps go up from the crowd in Times Square.

And in Britain, where they are apparently able to take these things with less commotion, @guardianstyle comments: “Early reaction to that #apstyle about-turn on email: ‘I say unto you, that likewise joy shall be in heaven over one sinner that repenteth.’ ”

No doubt other earth-shattering changes will appear this spring when the 2011 edition of the stylebook comes out. Until then, brown paper bags will be distributed to those hyperventilating over the initial announcement.

As a practicing copyeditor, McIntyre is entitled to his mocking hyperbole. I share his sentiments entirely; this is not even a tempest in a teapot, it’s a fuss in a thimbleful of spit, a matter of no consequence at all.

And, as McIntyre understands, the AP Stylebook (or Style Book, or whatever) has no scintilla of the force of law and is often profoundly silly in its blanket declarations. So the astonishing response that this change in advice has elicited on the web is dismaying.

Please do not write to tell me that my practice of spelling the word E-MAIL rather than EMAIL is now simply incorrect (or, worse, and incomprehensibly, “ungrammatical”) and that I should be publicly shamed. There are justifications for both spellings, and why should there have to be One Right Way, and anyway why should anyone care?

But let me take this out of the realm of complaint and counter-complaint and say some things about the general issue of spelling two-part expressions as separated, hyphenated, or solid. There are a great many cases, with a lot of subcases. Some (but not all) of them follow.

1. Compound nouns. Dining room, dining-room, or diningroom? Copy editor, copy-editor, or copyeditor? Short cut or shortcut? And many hundreds of other cases. In such cases, the alternative spellings are nothing more than that: the expressions are units lexically, morphologically, syntactically, and prosodically (with “compound accent”, primary plus secondary). Preferences differ in particular cases, differ for different speakers and communities, and change over time (so that the practice of dictionaries is forever shifting, though with some overall tendency against hyphenation and some overall tendency towards solidification). See cell phone/cellphone and smart phone/smartphone above. There is no substantive issue here, though a tremendous amount of heat is sometimes invested in particular cases.

E-mail/email is a related case, effectively a compound noun (including the accent pattern), though there is no separated variant.

2. Combinations with all. Most historical combinations with first element all are now standardly solid (and most have an unaccented first syllable): almighty, almostalone, already, also, although, altogether, always. These have new lexicalized meanings, with the contribution of all hard (or impossible) for modern speakers to appreciate. Fresh combinations with all ‘every one’ are then possible:

We are all one on this issue. We are all ready to leave. We will go all together, not one by one.

All over ‘finished; entirely’ has not gone the way of the others (though the non-standard solidified spelling allover is fairly common). Nor has all right ‘ok’ (used in this way since the early 19th century), versus all right ‘every one (of them) correct’, though the orthographic variant alright for ‘ok’ continues to carry the day despite the stigma induced by a barrage of criticism. (A topic for another posting.)

3. Combinations with prepositions. Many historical combinations with prepositions have been standardly spelled solid for a very long time: ahead, aloud, asleep, and others. The preposition + PRP verb combination (which is itself now non-standard) is, however, conventionally spelled with a hyphen: “I was a-swingin’ on the gate”.

There is currently variation in the spelling of for ever/forever ‘always’, though the solidified spelling seems to be gaining ground fast. And variation for in as much as vs. inasmuch as (though the handbooks caution you against in asmuch as).

And disputed usage on under way vs. underway. (Garner notes that solidification has been underway here for some time and suggests that it might be best to solidify the combination in all its uses.)

And non-standard, and stigmatized (though moderately common) aswell for the standard spelling as well, as in “I left, and Kim left aswell”.

4. Indefinite article plus head noun. Three notable cases here: stigmatized alot for a lot (as determiner or adverb); stigmatized ahold for a hold in get a hold of (itself a variant of get hold of); and disputed usage on awhile/a while. Each of these deserves a posting on its own.

[Back in comments on Ben Zimmer’s Language Log posting on “Zoological analogies” (here), the thread began with a query not about the content of the posting, but about the spelling awhile in a quote in the posting: “an invitation … to join the 1976 campaign for awhile”. This inspired a long rambling chain of comments on solidifications of several sorts, interspersed with comments on zoological analogies, though the solidification discussion predominated.]

5. Other indefinite determiner plus noun. This is a very complex domain, taking in combinations of any, every, some, and no with time, day, place, and where. Plus the closely related case of any more/anymore. Again, many of these deserve postings on their own. A few notes:

Brians’s Common Errors stipulates that every time is always separated, while anywhere, somewhere, and nowhere [also everywhere] are always solid. This accords with the distribution of else vs. other, according to which every time functions syntactically as two words (every other time, *everytime else), while the -where words function syntactically as single words (*every other where, everywhere else).

But both any time and anytime occur, with different accentual possibilities and different syntax (any other time, anytime else); similarly, some time and sometime.

And every day and everyday both occur, but they’re different accentually and syntactically (I go every day, an everyday occurrence).

Despite all this, the solidified spellings anytime, sometime, and everyday are common, even in uses that would otherwise call for separated spellings. Similarly, someday and someplace.

6. Some miscellaneous cases. Three cases that are standardly separated, but are increasingly spelled solid: never mind/nevermind, more so/moreso, thus far/thusfar. All three deprecated in Brians’s Common Errors.

Plus solidification of bare feet to barefeet, as in the beach sign No barefeet (presumably under the influence of barefoot as a modifier, as in barefoot boy). And related solidifications of back seat to backseat (sitting in the backseat) and back yard to backyard (working in the backyard), again from modifier uses, which are spelled solid and have compound accent.

Then there are sequences of words that are not in fact in syntactic construction with one another, but are sometimes incorrectly fused in spelling because of the influence of other spellings: into in giving into temptation, along in along night, aloud in aloud sound, forover in forover thirty years (under the influence of forever, presumably).

7. Fused of, have, and to. Finally, some familiar cases where an informal spelling represents reduced of, have, or to when they are phonologically (in some cases morphologically as well) fused with a preceding element:

a lotta/alotta for a lot of, lotsa for lots of
kinda for kind of, sorta for sort of
woulda, shoulda, coulda, mighta, musta, with -a for have/‘ve
wanna, gonna, hafta, useta/usta, with -a for to

No doubt there are still more cases. Solidification marches on.

12 Responses to “Solidification”

  1. Jan Freeman Says:

    “Then there are sequences of words that are not in fact in syntactic construction with one another, but are sometimes incorrectly fused in spelling.”

    And there’s the sequence often incorrectly UN-fused: “sing-a-long” for “sing-along.” But I haven’t figured out what model is influencing this one; “sing-along” seems so transparent I’m always surprised by the extra hyphen.

  2. Martyn Cornell Says:

    To cut a long story short, Jan – that, possibly.

  3. arnold zwicky Says:

    Follow-up from McIntyre here.

  4. Oliver Lawrence Says:

    To me, “e-mail” suggests an electronic variant of mail, whereas “email” is just, well, email, something in its own right as opposed to a sub-class or special case of something else, which is probably why I lean to the hyphenless (hyphen-less?) version.
    I also get the feeling that joining-up is more of an American than a British preference. Either way, what matters IMHO is that the writer identify what the “preferred” variant is for his/her genre and stick consistently to it. And keep a dictionary to hand…

    • arnold zwicky Says:

      “Either way, what matters IMHO is that the writer identify what the “preferred” variant is for his/her genre and stick consistently to it. And keep a dictionary to hand…”

      These are hopeless pieces of advice. Genres don’t have preferred variants, style sheets do (the NYT continues to insist on “e-mail”, while the AP now insists on “email”, and so on); and dictionaries differ, and change their preferences over time.

      As for consistency, this is one of those cases where readers are accustomed to see both variants, clearly without meaning differentiation, so that (unless they are committed to consistency for its own sake and are willing to scan through text checking for consistency in such matters) they are unlikely to notice variation even within the same text.

  5. Linda Lowenthal Says:

    Oliver, why do you think it’s called “e-mail” (or even “email” if you insist), then? Just coincidence? I am honestly baffled by this argument.

  6. The Raven Says:

    The problem with “e-mail” isn’t the hyphen, it’s that the “e” should be uppercased, per G-man, B-girl, A-bomb, etc. Rather than “email,” I’d prefer we just went with “mail.”

    • arnold zwicky Says:

      “Mail” will often do, as in “You have mail”, where the context makes things clear. But sometimes not; I get a pile of e-mail and snail mail every day, and often feel the need to be clear about which medium was the source of some message.

      It’s true that letter+noun used to be invariably spelled with the letter uppercased, but the practice has been slowly eroding, as in “t-shirt”, “x-ray”, “b-boy” (for “break boy”, a breakdancer), and a number of others. “E-mail” does occur, but not often, and spellings with the letter lowercased are clearly the norm.

  7. Oliver Lawrence Says:

    Linda, thank you for your robust comment; your perplexity probably indicates a lack of clarity in my earlier remark. My suggestion is that the name “e-mail” (obviously short for “electronic mail”) implicitly characterises it terms of regular postal mail, as a variant – i.e. an electronic one – thereof, whereas it has now become such a widespread feature of modern life as a communication vehicle of equivalent standing that it is worthy of a name in its own right. We’re all familiar with it. We know what the “e” is for. No need to draw attention to it with a hyphen any more.

    • arnold zwicky Says:

      I had hoped to avoid a descent into a back-and-forth about the virtues of one spelling or the other. Proponents of each spelling have good reasons for their choice, so there’s absolutely no point in people bashing other people’s choices and insisting that there must be One Right Way.

  8. Complaining « Arnold Zwicky's Blog Says:

    […] digression: The solidified spelling illusage looks decidedly odd, and in fact OED2 doesn't otherwise countenance it, giving […]

  9. Separated spellings « Arnold Zwicky's Blog Says:

    […] orthographic words) or solid (as a single orthographic word)? (See discussion of solidifications here.) Two examples — of different types — that have come to my attention recently: for ever […]

Leave a Reply

%d bloggers like this: