Michael Hartford

Caleb Crain has an interesting little rant in the New York Times Magazine about the practice of “camel casing”: inserting a capital letter in the midst of a (usually) compound word, with or without an initial capital–MasterCard, iPhone, PowerPoint. Crain is dead set against it, though more because it so often represents a mangling of language by corporations than because of its effects on readability. Indeed, most of the piece is about the practice of putting spaces between words–a comparatively recent innovation, according to Crain, developed by Irish and English monks in the 13th century to aid in the transcription of Greek and Latin manuscripts–and not really about camel casing at all.

Interestingly, Crain alludes to the practice of camel casing in two language areas with which I have some familiarity: Irish Gaelic and computer programming. In both domains, the camel casing adds significantly to reading comprehension; whether it also aids comprehension in other domains (like trademarks, which is Crain’s major beef) is perhaps open to debate.

In Irish, it’s a useful convention because words often change their initial letter depending on grammatical case. Crain uses the example of “Myles na gCopaleen,” one of the many pen names of Brian O’Nolan (also known as Flann O’Brien and Brian Ó Nualláin). “Copaleen” is a typically O’Nolanesque transliteration of the Irish “capaillín,” or “ponies” (literally “little horses”). Because he is Myles of the ponies, “capaillín/copaleen” is in the genetive case, so the initial “c” is replaced by a “g”. (Trust me, this makes perfect sense; it’s called “eclipsis,” or “urú” in Irish.) To make it easier to see the root word in the transformation, the practice in written Irish is to capitalize the original initial “c” in proper or otherwise capitalized words, hence “gC.” The same practice can be seen in the song title “Casadh An tSugain” (“Twisting of the Rope”), and the pagan paradise “Tír na nÓg” (“Land of the Young”). “tSugain” is significantly more readable than “Tsugain” would be.

When not used in a proper or otherwise-capitalized word, it is common to see a dash placed between the eclipsing letter and the root word: “na g-capaillín,” “an t-sugain,” “na n-óg.” Again, readability is the goal. The eclipsing letter is really more like a diacritical mark than a letter; it changes the pronunciation (the “g” in “g-capaillín” is pronounced, the “c” is not), but does not really change the word’s spelling. (In classical Irish script, a dot beneath the letter was used to indicate lenition, the other Irish initial mutation, rather than the following “h” used in modern Irish: “mac/mo mhac”.)

The other (more common) domain in which camel casing is found–indeed, where the term itself is frequently used–is in computer programming. The practice arises from the need for human eyes to easily see and understand the words (or “tokens”) in a program: it is much easier to read and comprehend “ApplicationConfigurationManager,” for example, than “applicationconfigurationmanager.” The computer, of course, couldn’t care less about the capitalization patterns (so long as they’re consistent: in many languages and on some platforms, the two tokens above could refer to entirely different objects). Because one of our first responsibilities as programmers, as I’ve noted elsewhere, is to be helpful, camel casing is an almost unconscious behavior among most developers.

The other camel-case pattern that Crain decries, with an initial lowercase letter (like “iPhone”), is also common in programming, at least among the older set who had to deal with loosely-typed, late-binding languages without the aid of modern programming tools, called “Hungarian notation” after the work of Microsoft chief architect Charles Simonyi. In this notation style, variables contain their type–text, integer, etc.–in their names, as the initial part of the name. For example, “strName” indicates that the variable is a text string; “iCounter” or “intCounter” indicates a numeric integer. This habit is a bit redundant in many modern languages, where confusing an object’s type is an error detected by the compiler, and with many modern programming tools, where one need only hover the mouse pointer over a variable name to see what its declared type is. But anyone who’s had the experience of coding in a plain text editor (I wrote my first Java in the UNIX “vi” editor, and also walked uphill both ways in the snow barefoot), or worked in a late-binding script language like JavaScript or PHP, will appreciate this extra indicator of what’s supposed to be happening. Linus Torvalds may not like Hungarian, but I do.

Crain, of course, isn’t concerned with how Irish writers or computer programmers handle capitalization; these are niche applications of language at best. Indeed, his beef isn’t exactly with camel casing per se; rather, it’s with the efforts of corporate branders to make their names stand out, and his column is a call to subvert the subversion:

Perhaps camel case, like intrusive music, baffling floor plans and aggressive fragrances, is deployed to weary and bewilder us, to render us so addled that we have to say corporations’ trademarks aloud to be sure of what we’re looking at. It doesn’t have to be this way. Put some distance between you and your Master Card; don’t let your Iphone make the rules. You don’t have to buy their language. It already belongs to you.

The reference to “say[ing] corporations’ trademarks aloud” refers to the part the medieval innovation of letter spacing played in permitting silent reading; when words are run together, either in the ancient scriptura continua fashion or in the contemporary branding camel case, sentences need to be voiced to be understood. By forming their brands in this fashion, Crain suggests, corporations are forcing us to vocalize their names, which at the very least slows us down and also, as in some ancient magical rite, implicates us in invocation: spoken names have power.

To a great extent, I agree with Crain, at least in sentiment: preventing corporations from colonizing our language (who already would like to reduce many common words–like Word itself–to trademarks) is in all our interests. The branding assault is rarely as subtle as camel-cased brand names; and that’s where I can’t quite get on board with Crain’s suggestion that the camel-casing trend is quite so sinister as a sort industry scriptura continua aimed at making us slow down when we see their names. Indeed, once we become used to the shape of their brands on the page or screen–iPhone, PowerPoint, MasterCard–we don’t slow down at all; they become sight words, smoothly slipping into the subconscious (which may be more insidious than a plot of forced invocation).

The camel-casing trend seems to me to be just that: a trend, not unlike the fake-name trend (Qwest? Cingular? Ameriprise?). The camel cased name has that Web 2.0 feel, with its roots in the technology world (I doubt they’re going for the Gaeltacht cachet), even if the branders have no idea why the pattern began in that world. Indeed, camel casing is prevalent in the decidedly non-corporate open source world: SourceForge, FileZilla, OpenOffice. DaimlerChrysler, FedEx, ExxonMobil, and GlaxcoSmithKline look more like middle-edged men with combovers trying to look like the cool kids than anything particularly (or at least effectively) sinister.

Like any other trend, I expect the camel casing to come to an end when so many camel-cased names exist that they no longer stand out. Then something new will be necessary: weird punctuation (like Yahoo! and Guess?), bad spelling (like Qwest, to whom I still refuse to make out a check: they will be U.S. West in my Quicken settings until they cut off my telephone), and faux foreign names (Häagen-Dazs) are ripe for plunder. It’s the rate of name changes–useful either for keeping a brand “fresh” and “in the news,” or for trying to dodge an unsavory history (hello, Blackwater Xe!)–that is more interesting, and perhaps more sinister, than a transient fad.