Archive | December, 2015

Hits and misses

21 Dec

Perhaps too there will come along soon as ingenious an individual as a young clerk of a large New York firm that used mailing lists for circularization purposes and found it difficult to decide what prefix to use before a woman’s name when there was nothing to indicate whether she was married or single …

This bright young clerk solved the difficulty in so simple a way that it is a wonder that nobody ever thought of it before – by a compromise, the means of settling difficult and disputed points ever since the world began. He used the prefix “Ms.,” equally applicable to married and single ladies. His salary was raised in consequence.”
(Arizona Republic, 22 December 1922)

Linguist Ben Zimmer’s research into the earliest days of “Ms”, published last week on Listserv, uncovers a fascinating story of progressive prescriptivism in action – not the kind of prescriptivism that tries to keep new words out of the language, but the kind that tries to get them in.

As he recounts, the word was unabashedly fabricated to offer a new solution to an old problem: one that was exercising polite society even as far back as the 19th century. The first two appearances of “Ms” in print are both in the form of straightforward, prescriptivist proposals to avoid social embarrassment. In 1901, the Springfield Republican suggested:

“What could be simpler or more logical than the retention of what the two doubtful terms [Miss and Mrs] have in common? The abbreviation “Ms.” is simple, it is easy to write, and the person concerned can translate it according to circumstances’.

Then, in 1912, the New Orleans Times-Picayune, as an aside while discussing the suitability of singular “they“, said:

A more serious need is a common title for Mrs. or Miss to be used in cases where the status of the person spoke of is unknown or irrelevant … In such a case Ms would be at once neat ambiguous and appropriate.

Then, nervously at first, the idea catches on.  In 1921, in a submission to the Cleveland Plain Dealer, Ann Taylor writes with an anecdote:

An advertising agency wondered how it would address letters to a list of women whose names bore no indication as to whether they were “Miss” or “Mrs.” The No-Less-Than-Authority, the President of Harvard, informed them that it is quite correct if in doubt, to use the prefix “Ms.”

Then a syndicated columnist writing from New York, Lucy Jeanne Price, says:

It is a constant complaint that women no longer add “Miss” or “Mrs.” before their names, and that consequently in writing a business letter to a strange woman, one never knows how to address her. One New York firm has solved the problem by the ingenious adoption of a telescoped prefix, “Ms.”

And the story of the advertising agency catches on, as the Pittsburgh Dispatch relates:

One large New York firm that uses mailing lists for circularization found it difficult to decide what prefix to place before a woman’s name when there was nothing to indicate whether she was married … Finally a bright chap suggested the prefix “Ms.” As a hedging scheme this worked fine. The clerk who made the suggestion received a raise.

Understandably for an out-and-out neologism like “Ms”, a newspaper editorial is not quite enough to persuade early adopters: they are eager for some authority (the president of Harvard), or at least precedent (the mailing clerk) to justify its use. But with an academic and a New York corporation leading the way, there is enough cover for the idea to take root. Later, as related by the philologist Mario Pei, the suggestion found its way into advice and usage guides, the body of prescriptivist authority grew, and a courtesy title was born.

Of course, prescriptivism’s record on innovation isn’t always so successful. For example, the many inventive proposals for an epicene alternative to “he or she” – heshe, hizzer, heesh et al – all foundered in the face of singular “they”. But this is a slightly different case: the neologisms were being proposed not because there were no alternatives – singular they has been in use for 500 years – but because there were objections to the aesthetics or syntax of using a plural pronoun in a singular sense. There was no actual need for a new word; just a desire for a more euphonious one.

In an excellent post earlier this year, Tom Freeman, The Stroppy Editor, suggested that English might be best understood, not as a code or law, but as a kind of linguistic free market:

The people best placed to judge clarity are the people. All of us. We are the ones reading, writing, talking, listening. If we find that one way of using words helps us communicate more clearly than another, then we will favour the better way … Language is controlled by market forces. Nobody is in charge, and that makes our language far more dynamic, efficient, and rich.

The analogy works perfectly to describe the way language functions: it is pointless for conservative prescriptivists to stand in the way of popular demand, or offer new “products” to customers who already own a better alternative.

But sometimes, linguistic demand exceeds supply. The slow, contemporary emergence of “Mx” as a gender-neutral courtesy title is a good example: in this case, it’s not that there are a selection of controversial or ugly alternatives; it’s that there are no transgender honorifics in English at all.  The market in words has failed. So “Mx” – proposed, just like “Ms” a century go, as an unashamed neologism to fill a gap – has found its way into the Macmillan and Oxford dictionaries and (who knows?) may find a slew of interested customers.

Descriptivist linguistics, which observes rather than intervenes in language, is the best model for studying this market. But economists are not inventors, and those who chart the language as they find it can be blind to the words that aren’t there. Sometimes, you need a new coinage from a commentator, a campaigner, a usage guide writer; an invention, if you like, from an entrepreneur. That’s prescriptivism’s chance to shine.


Accents will happen

7 Dec

If you think putting all the accents on Société Générale is a bit of an effort, imagine picking up a story that combines Turkish with Middle-earth:

Picture 15

For years, the software at the Tribune offered only the English alphabet plus the common western European diacritical marks (acute, grave, umlaut, tilde, etc) for editing purposes. But the latest upgrades now offer a splendid range of glyphs, allowing us to range typographically much further into the east and south of the continent. The only question now is: how much should we use them?

The Tribune’s policy for a while was simply not to put accents on any foreign word at all – a bald, if even-handed, approach that raised eyebrows with readers and caused anguish among the multilingual members of the subs’ desk. That policy was subsequently modified to a more common and pragmatic approach commendable for its humility:

accents Use them in French and German words that take italic. In foreign words that have become part of the English vocabulary – cafe, cliche, detente, denouement, debacle, protege etc – no accent is required …

Keep accents on proper names. Thus: ‘Arsène Wenger was on holiday in Bogotá with Gérard Houllier’ … In general, other marks such as Turkish should be avoided for fear of introducing errors as much as difficulty in checking

Now, however, with the full expert character set at our disposal, the policy is to go for it all: every unfamiliar word and name rendered as it should be in its native language. That, you would think, is the definitive, unarguable solution. But even this approach – assuming your font library can stand it – raises one or two questions.

The total-accuracy policy does not extend, of course, to languages with entirely non-Roman alphabets – Russian, Greek, Chinese, etc. English-language newspapers don’t expect their readers to be able to understand them. But several Turkic and Baltic languages have many characters that are not classically Latinate and are certainly unfamiliar in English. Turkish, for example, has a 29-letter alphabet, including six letters that don’t appear in English: ç, ğ, ı (a dotless “i”), ö, ş, ü. Their sound and function is a closed book to many general readers. Doesn’t that make them as much candidates for transliteration as π or Σ in Greek?

There are pros and cons for all three of these approaches, as set out below. On balance, I’m glad we’re going for (3), even though there’s obviously a risk we may end up with a broad-based version of (2).

And once I’ve checked all the Turkish, it’ll be time to tackle Hobbit nomenclature!


(1) Removing accents and foreign characters completely

• Impressively and genuinely consistent
• Avoids the slightly uneasy relativism of approach (2)
• Easy to remember and enforce

• Every multilingual reader will see, and frequently object to, what you’ve done
• Faint hint of insularity (don’t they speak any languages?)
• Theoretical risk of confusions over near-homonyms that differ only by a diacritical mark or character  (but, then again, only to readers who speak the language)


(2) Putting in accents and foreign characters for locally familiar languages only

• Relatively easy to remember and enforce
• Fits quite well with your audience’s likely linguistic knowledge
• Possibly as much as your technology can reliably achieve

• Which languages are actually “commonly spoken” in your area? Or are you just paying attention to the ones the chief sub speaks and ignoring the others?
• Can be unconsciously revelatory of the limits of your expertise, or at least your horizons
• Can look very awkward when words in a language you do put accents on appear in the same story as words in a language you don’t


(3) Putting in accents and foreign characters for all Latin-alphabet languages

• The gold standard for accuracy
• A genuine editing challenge
• Impressively authoritative and literate

• May be more than your software can manage
• Finer distinctions may be lost on, or intimidating for, readers
• Are you sure you’ll always recognise when there are characters or diacriticals missing? How’s your Lithuanian?*


*32 letters, 5 digraphs