When someone on the phone—the doctor’s office, the bank, the credit card company—asks for my name, I always offer to spell it out—it’s a pretty uncommon surname. So far as I know, there are somewhere between 10 and 20 Nosowitzes in the world, and they’re all closely related to me. Because it’s uncommon, and because it would be a problem if my bank writes my name down as “Moskowitz,” I err on the side of caution. “N as in Nancy, O, S as in Samuel, O, W, I, T as in Thomas, Z as in Zebra,” I chant.
This uses what is what’s called a “spelling alphabet,” or, confusingly, a “phonetic alphabet.” (The latter is confusing because it has little to do with phonemes, or a unit of sound in a language. Plus there’s the International Phonetic Alphabet, which is something else entirely.) The history of spelling alphabets is fascinating and winding, but it’s notable that there hasn’t been an official update to the most commonly used English version in about half a century. We might be in need of one. As mobile phones have replaced landlines, call quality has, strangely, gone down. The general connectivity of the world—including the ease of international video calls and the use of foreign call centers—means that spelling out a name or word is an increasingly common practice. A modern, updated, globally friendly English spelling alphabet would be pretty useful right now, but getting people to use one might be harder than you’d think.
All alphabets that use letters to represent sounds have names for their letters, to refer to them for things like spelling. English letters fall into rough groups for these purposes. Vowels are usually pronounced with their long sound. Stopped consonants—those that force you to halt the flow of air from your mouth, like B, D, and T—are named as if you wrote “ee” after the letter: “bee,” “dee,” and “tee.” Nasals (sounds made through the nose, like M and N) and fricatives (sounds made by forcing air through a narrow space in your mouth, like S and F) are named by putting the letter “e” before them: “emm” and “ess.” Then there are the weirdos with their own histories, like J, Y, and Q.
This is why the names of most letters sound an awful lot like some other letter. Taking the Nosowitz example, it’s easy to confuse “zee” with “cee,” or “ess” with “eff.” This wasn’t a huge problem for most of history, when humans who needed to spell things out were generally standing pretty close to one another. But when people began to gain the ability to communicate over long distances, problems arose.
“The story goes that it started with semaphore relay stations,” says Brian Kelk, a computer scientist who’s worked at Cambridge University and who maintains an exhaustive page about spelling alphabets. He’s not a linguist or anything; he just says he happens to be fascinated by this stuff. “Someone would be watching incoming signals and shouting out letters to someone sending outgoing. They invented phonetics for some troublesome letters,” he says. The British military came up with the first few examples, just for letters they found the most difficult: “P as in pip,” “B as in beer.” Those were set down in regulations in 1904. Between then and the end of World War II, the British, the Americans, and various telecommunications companies kept working on these alphabets, producing dozens of standards.
Just as the names of letters are in groups, spelling alphabets also might have some sense of thematic organization. “We remember better things which are linked in terms of their meaning,” says Valerie Hazan, a professor of speech sciences at University College London. (We’ll be hearing a lot more from her.) Geographic place names are one group. From the 1912 Western Union Spelling Alphabet, just for example, we have “B as in Boston,” “N as in Newark,” and “T as in Texas.”
First names are another group. From the 1917 Royal Navy telephonic alphabet: “G as in George,” “W as in Willie,” “E as in Edward.” And, bizarrely, dances show up as a group: “F as in Foxtrot,” “T as in Tango,” “J as in Jig.”
For about 80 years, governments and corporations futzed with these spelling alphabets, and learned that some stuff didn’t work—it turns out, for example, that “Lima” is also the Malay word for the number five. A tremendous amount of research, time, and money was invested into figuring out the optimal spelling alphabet—at least for the three languages that the International Civil Aviation Organization (ICAO, the United Nations agency that handles air transportation) felt significant enough to have one (English, French, and Spanish). The ICAO scrambled, using researchers across the globe on the problem, and by 1959 had finalized what is today probably the best-known spelling alphabet: Alfa, Bravo, Charlie, Delta, Echo, and so on. (As a side note: “Alfa” is not a typo. The whole “ph equals f” thing is confusing, and reasonably so, for non-English speakers. The same goes for the alphabet’s J—Juliett with a doubled final letter so the French won’t say “Juliay.”)
A tremendous amount of research, time, and money was invested into figuring out the optimal spelling alphabet.
That is the now the standard alphabet for organizations including NATO (which often lends its name to the alphabet), the Federal Aviation Administration of the United States, the International Amateur Radio Union, and pretty much any international group that wants or needs a standard. It’s certainly the most commonly used spelling alphabet in the world, but it is, as most of these alphabets are, exceedingly Anglocentric.
Other languages have come up with their own spelling alphabets. Some needed wholly new ones, such as Russian, which uses the Cyrillic alphabet. “Г as in Григо́рий” is the Russian version of “G as in Gregory.” Japanese and Mandarin Chinese both have their own letter-based alphabets (Kana and Pinyin, respectively) in addition to their traditional logographic alphabets (in which symbols stand in for whole words or phrases, rather than just sounds). Those letter-based alphabets, in turn, have their own spelling alphabets.
Some languages that use the Roman alphabet, as English does, have letters of their own. Take Æ in Danish and Norwegian, which is usually given “Æ as in Ægir,” a figure from Norse mythology. Spanish has the Ñ, though precious few words start with it. Despite that, most people go with “Ñ as in Ñoño,” which means “dull.”
English was the first language to have these spelling alphabets, so perhaps that’s why other languages follow its general outline and themes. Most spelling alphabets around the world are still based on place names and given names: (the equivalent of) “と as in Tokyo,” “Δ as in Demetrios,” or “И as in Ivan.”
But by around the late 1960s, things had calcified. The official or widely accepted spelling alphabets were set. Yet here we are, in 2019, with an actual need for a new one.
Voice call quality has gone down over the past two decades. Mobile phones have added convenience and a million other things, but they have done away with a wired network dedicated solely to voice communication, as well as the large microphones and speakers of old landlines, which featured decades of refinements to improve call quality. Cell phones, on the other hand, rely on tiny, awful microphones, and tiny, awful speakers for calls, and tiny, awful allocations of bandwidth. Mobile networks can be compromised by everything from streaming video, to the presence of a tree or wall, to the weather. Most importantly, they’ve dropped the emphasis on voice quality. Nobody seems to care any more whether they can hear a pin drop.
Sure, much communication has moved over to text, email, and social platforms, but everyone still needs to talk on the phone sometimes. English is such a widely spoken language that it comes in many different flavors, dialects, and accents, which further complicates understanding people clearly over a shoddy mobile connection. Independent of their use in military and aviation capacities, we sort of need spelling alphabets now more than ever. The problem is that what we’ve been given by the 50-year-old standard is deeply flawed for modern use.
“We know in speech perception that frequent words are much more easily heard in noise than infrequent words,” says Hazan. That why it is a pretty poor choice to use, say, “S as in Sierra” (the standard) instead of “S as in sugar.” And many of these alphabets were designed for international use. So why do some of them use hyper-local words like “Newark,” or strangely outdated, unfamiliar names like “Ida,” “Bertha,” or “Nelly”?
The options for certain letters—Q and X, we’re looking at you—are not great, but come on. “Q as in Québec” is a truly awful choice, with multiple pronunciations (“keh-beck,” “kweh-beck,” and I even heard a Brit in a radio interview say “kyoo-beck”). And the emphasis is on the second syllable, not even on the syllable with the most important letter! Early common choices for X included “Xanthippe” (the wife of Socrates or some figure in Greek mythology?), “Xaintrie” (a town of 300 in central France), and “Xerxes” (a couple of Persian kings). Eventually everyone settled on “X-ray,” which is fine, but basically the equivalent of saying “X as in X.”
Despite the alphabet needing to have global use, some of the words are incredibly difficult to pronounce for some people. Think about the word “foxtrot.” “That’s virtually impossible to say,” says Hazan. Right in the middle is a traffic pileup of consonant sounds: K, S, T, and R, with no vowel break. For speakers of Spanish, which does not allow so many consonants to be grouped together without a vowel, it is an insanely difficult word to say, not to mention about as outdated as the name “Bertha.”
Hazan, in 2006, was asked for a BBC Radio story to see if she could come up with a better spelling alphabet. “We chose 12 words to be replaced, the ones we felt the most problematic,” she says. “Alfa” became “apple,” “Québec” became “queen,” and “Zulu” became “zebra,” to name a few. These selections were informed by common, easy-to-pronounce words that can carry through noisy connections. (Soft sounds, such as the “th” in “thought,” are much lower in volume and harder to hear than, say, a vowel.)
To test her plan, Hazan set up noisy phone calls, and tried to convey four-letter nonsense combinations (like, say, “Apple Zebra Queen Lemon”) to both native and non-native speakers of English, in both her new alphabet and the old ICAO standard. “We put quite a lot of work into this, and were quite certain that our new words were going to be the ones to win out,” she says.
So, these words corrected for all of these problems of localization and soft sounds and outdated terminology. They must have been much more effective, right?
“No, they were not,” says Hazan, with some dry amusement.
People tend to make up their own spelling alphabets and pass them down.
Turns out there was effectively no difference between the new, improved spelling alphabet and the old standard. If certain letters were in certain places in the nonsense combination, the new version might be more effective; in other places, the old version was. No difference! After all that!
This can be partly explained because people have just grown familiar with the whole “alfa-bravo-charlie” thing. It’s in books and movies, it’s just one of those things we absorb without thinking about it. The other possible explanation is that these are equally effective or ineffective because people tend to make up their own spelling alphabets and pass them down. Anything beyond what you’re used to is the same as anything else, more or less.
The way I spell my name is one of these personal versions. “N as in Nancy” appears in a 1947 ICAO attempt (though today it is “November”), but “S as in Samuel” is nowhere. My version is cobbled together. “Nancy” comes from my dad, I think. I have a vague recollection of coming up with Samuel on my own—though I’m sure I must have heard it somewhere. It seems to work, so I’ve stuck with it.
The ICAO or NATO spelling alphabet might be by far the most common one of its type, but I’d bet it’s nowhere near as popular as the millions of individual spelling alphabets we all devise and pretty unknowingly pass to our children. And here’s the real problem of coming up with a new standard: Everyone who needs one already has one, and it’s whichever of the millions of variations is most comfortable. It’s absolutely possible that “Sierra” or “Susan” or “Susquehanna” would be a better choice than “Samuel,” on the scientific basis of difference from other words or frequency of phonemes best suited to a noisy cell phone connection. But I’m sticking with “Samuel.”