I speak in the low grunts of a bachelor, so sometimes I have a hard time being understood in public places with a lot of background noise. It’s especially problematic in the coffee shops of New York City, where conversation and clinking saucers make a dull roar. In American coffee shops, cashiers often ask for a name to go along with each order, especially at peak hours. Doing so mostly serves to differentiate among the six small lattes that were ordered in the previous ten minutes, as names are called out once each drink is prepared. Thanks to the difficult environment, no doubt, it’s not uncommon for names to be misheard, misspelled, and mispronounced, a phenomenon ubiquitous enough to be lampooned in popular culture by the likes of How I Met Your Mother and Saturday Night Live. Democratized collections of the most unusual of these misunderstandings come in the form of Tumblr blogs such as Starbucks Spelling.
I typically manage to place my order smoothly enough, but cashiers always ask me to repeat my name. After I half-shout “Greg!”, enunciating as well as I can, they respond with a bored shrug and scrawl their best guess onto the cup. I’ve retrieved many drinks from the counter with “Craig” or “Rick” or “Grey” written on them, but they are all preferable to some of the peculiar non-names that I’ve been given, like “Rank.”
Tired of this repeated miscommunication, I decided to design a “coffee-cup name” specifically for use with baristas and cashiers. I don’t want just any name. I need the perfect pseudonym. Inspired by a collegiate dalliance with linguistics, I have a gut feeling that some names are easier to hear because they’re made up of sounds that are easier to hear. The ideal sobriquet à café will combine those sounds to successfully pass through each stage of the coffee-shop game of Telephone: from me to the barista one time without repetition, then written flawlessly onto the cup, announced perfectly when the drink is ready, and finally heard by me through the din.
Drawing from a limited inventory of easy-to-hear sounds should narrow the field of possible pseudonyms, but only one of them can be the easiest to understand in a coffee shop. I intend to find that name and make it mine.
José R. Benkí is a research investigator in the Survey Methodology program at the University of Michigan. He knows a thing or two about interpreting messages in noisy environments, otherwise known as speech intelligibility. His research has looked at the confusability of English phonemes, which is assessed when subjects listen to stand-alone nonsense syllables (e.g., /bep/ or /tid/) in different levels of noise and transcribe the sounds they can hear. In the absence of noise, all speech sounds are easy to hear. When the noise levels rise, recognition varies. The sounds that are correctly transcribed more frequently are more identifiable, and the rest are confusable to varying degrees.
I reached Dr. Benkí over the phone, but in an example of poor intelligibility, he had trouble understanding me. After we switched to VoIP, he first wanted to make it clear that there are two problems when it comes to understanding a name or any other message in a coffee shop. There’s the matter of the identifiability of the sounds themselves, which is the reason I sought out his advice. Then there’s the matter of what the listener expects and how predictably the speaker speaks. I pressed him to explain which sounds should be the easiest to hear in a noisy coffee shop.
“Consonants are by their nature short, quick, and transitory,” he began, explaining that some consonants are more identifiable than others. The most identifiable consonant is /s/, which is a loud, continuous noise. By contrast, some consonants are produced with a brief blockage of airflow, which produces a period of silence that might be obscured by noise. Still, among those consonants, /p/, /b/, /t/, and /k/ were more often correctly transcribed in his experiments. Dr. Benkí said to avoid the “th” phonemes (transcribed as /θ/ and / ð/) and /m/ and /n/, which are very easily mistaken for one another.
I asked about syllable structure. “Consonant clusters can be difficult,” he said, although he didn’t study them specifically. Thus, the /r/ in “Greg” is confusable because it overlaps with other sounds. In a common consonant-vowel-consonant syllable, final consonants are more confusable than their initial counterparts. “This is all potentially useful information!” I thought, trying to make a list of names compatible with everything he’d told me.
Dr. Benkí recommended that syllables of my coffee-cup name start with a consonant, rather than a vowel. Then he warned me that even if a final consonant is not present, an American listener might perceive one to be present because these are so common in English; it is what the listener expects to hear. I was reminded of his second point from the beginning of the conversation. While the ideal coffee-cup name might happen to be very identifiable based on sounds alone, I should really be looking for the name that is the most intelligible.
Dr. Benkí’s phonetics experiments were meant to establish a baseline of identifiability for English phonemes with as little context as possible. Outside of the lab, his results may still help predict the intelligibility of a message, but listeners’ expectations play a crucial role in determining how sounds are interpreted. This is a good reason why I am probably better off with a pseudonym that is recognizable as a name. While a phrase like “baseball bat” might have sounds that are easy to hear, in a coffee shop it would definitely be unexpected. Context helps us make sense of words with confused phonemes, often trumping purely phonetic principles. In a lab experiment, /g/ is more likely to be mistaken for /d/ than /k/, but in a coffee shop, “Dreg” and “Red” are relatively uncommon names. Instead, my name is often mistaken for “Craig” or “Rick.”
There’s something else at work here. People can report hearing a sound clearly even though the sound in question was obscured by noise. Replacing noise with a phoneme—called the “phonemic restoration effect”—occurs because our brains fill in sounds we expect to hear. How is this relevant for a coffee-cup name? It might be an argument for adopting a longer name. A name like “Jonathan” is more resilient than “Jon” if one of the sounds is obscured, because its surviving sounds provide enough information for the listener to subconsciously fill in the blank correctly. (On the downside, longer names take slightly longer to say and write. They may be less common, and thus less recognizable, or more unexpected.)
I asked Dr. Benkí the most important question: “What name would you suggest?”
Dr. Benkí declined to suggest any particular pseudonyms. Names are too important to him to encourage anyone to choose a new one simply to avoid misunderstandings in coffee shops. He laughed explaining himself. He couldn’t take the idea seriously.
What would a real barista say about a fake name? A couple of days later, I was nursing a cold-brewed coffee at Joe in Grand Central Terminal, tucked away from the bustling Main Concourse. This location of the popular chain sees heavy traffic during rush hour, but it was also pretty busy on a Saturday afternoon even though there was no seating—not even a countertop to rest my drink on as I waited to meet with the manager, Sarah. I had spoken with her earlier on the phone.
Eventually she descended the spiral staircase in the back room, languid and lank, with the complexion of an iced coffee with milk. To my mild embarrassment, “Sarah” was actually Farah. I apologized for the confusion, but I was secretly glad to have an example of what I wanted to speak with her about.
Farah said that she asks customers to spell their name if she has a hard time hearing them, but not to avoid a misspelling. She sometimes writes names phonetically for the benefit of the staff even if it isn’t how she thinks the customer would spell it themselves.
“The important thing,” she said, “is to know their name.” She means the regular customers. A name is part of the personal interaction of coffee service, not just a label on a cup in a commercial transaction. By her estimate, 80% of the customers at this location are regulars, commuters who work in Midtown, and she knows most of their names.
I wondered silently, “Would a fake name make me an impersonal, inhuman monster?” For a moment, I felt like a real asshole.
Farah also said that cashiers appreciate any effort by a customer to be more easily understood, but I should identify with the name I choose, especially if I want to return to the coffee shop, because I am going to develop a relationship with the staff. A nickname or initials perhaps? She shared my intuition that longer names are generally more intelligible, and asked if my given name is “Gregory.” It is. I could use that, she suggested, not knowing that only my mother calls me “Gregory,” and only when she’s upset with me.
When I was talking with Dr. Benkí, he advised me to consider that our society, which makes up the cultural and phonetic context in which names are intelligible, is increasingly diverse and multilingual. This is why he has given his children names that will be properly pronounced and easily understood by native speakers of both English and Spanish. My coffee-cup name should be similarly flexible.
But as I saw at Grace Street, a surprisingly spacious coffee shop in the heart of Manhattan’s Koreatown, the cultural context is dynamic. A small sign on each table informed customers, in English and Korean, that the café would be switching to a “dessert café” format in the evenings. Grace Street is acclaimed for a Korean street food called ho-dduk, which they describe to neophytes as a “Korean-style doughnut,” filled with cinnamon sugar and served à la mode. Business must be booming, because as I sat sipping a latte, a deliveryman rolled in a dolly loaded with giant cylinders of vanilla ice cream.
By all appearances, the clientele that day was predominantly Korean or Korean-American, like employee Rachel B. She told me that these Korean customers tend to use their last names, like Kim and Park. “Very identifiable!” I thought. She also noticed people using what seem to be fake names. Popular characters in Korean dramas show up as pseudonyms. When a series called The Heirs was at the height of its popularity, the name of one of its central figures, Rachel, seemed to be in more frequent rotation. Rachel B. paused, then assured me that Rachel is, in fact, her real name.
She had often heard the name “Jay,” which she suspects is usually an assumed name. In fact, she has overheard groups of young Koreans teasing each other after one of them places an order as “Jay.”
“That’s not your name!” one might say.
“Shut up! It’s easier this way,” comes the reply.
“Easier this way” seemed to be a theme. When I suggested that Korean customers might be trying to take advantage of the cachet that comes from using their family name in a seemingly Korean place of business, Rachel B. did not agree. Whether using a last name or a pseudonym, she thought people go for something quick and easy, something they think will be understood.
After a dapper middle-aged foursome stepped to the counter and ordered in Korean, I asked Rachel about the “Korean-ness” of the place. When the coffee shop first opened, the staff was entirely Korean-English bilingual, but non-Koreans have joined their ranks. Now Grace Street may not be “Korean enough” for some of the patrons, Korean nationals who expect a Korean café experience when they are in K-Town. That means milk and sugar added by the staff in a magic ratio, and interacting in Korean, including the name on the cup. It’s not that way at Grace Street, where the milk and sugar are self-serve, and the cashiers aren’t always Korean. This generates some friction. Occasionally patrons refuse to talk to, or even acknowledge, English-only cashiers, instead speaking directly to Rachel B. behind the espresso machine.
As the culture of a coffee shop changes, so does the context for names.
When my conversation with Dr. Benkí was drawing to a close, I attempted to put everything together. “So you’re saying . . .” There was an awkward silence. I didn’t know what he was saying. I was hoping he would just tell me what name I should use. My conversations at coffee shops had further informed my search, but hadn’t supplied a specific pseudonym, so I decided to speak with more linguists.
As I made my way to the New York University Linguistics Department, “Van Lingle Mungo,” a song with the names of baseball players for lyrics, serendipitously began playing on my MP3 player. I stopped to admire the fountain in Washington Square Park as Dave Frishberg crooned, “Heeney Majeski, Johnny Gee / Eddie Joost, Johnny Pesky, Thornton Lee...” I still needed a pseudonym.
Lisa Davidson, director of the Phonetics and Experimental Phonology Lab at NYU, wrote me that coffee and phonetics are two of her favorite things. I believe her, since she discusses vowel mergers with the acuity and alacrity of the well-caffeinated. We were joined in her book-lined office by Sean, a graduate student in linguistics with long hair and glasses, whose five o’clock shadow was a little early. Neither of them had ever experienced any difficulty with their own names in coffee shops (it must be the sibilance), but they were interested in my search nonetheless.
We reviewed what I had learned. Some sounds are more confusable than others, but there are many confounding factors to account for when selecting an intelligible name. Name length, the perceived “actual name-ness” of a name, its frequency or rarity, and the number of similar-sounding names are interrelated pieces in the puzzle. Spelling and the connection to identity are also concerns, but their importance seems to be a matter of personal preference.
In principle, I could devise an algorithm for ranking pseudonyms, combining “phonemic confusion matrices” and baby name ranking data with similarity scores to produce a list of ideal names. Dr. Davidson seemed particularly excited at the prospect of all this number crunching, and encouraged me to take an empirical approach. She was much more optimistic about my chances of completing the computation than I was. Because I can hardly keep track of the variables, much less assign them numerical weights, there will be no massive dataset, no lines of code, and no empirically ideal coffee-cup name for me.
Still, the linguists were confident that acceptable options exist. They told me to look for high “name-ness” without any near neighbors—other names that differ by only one or two sounds. They agreed that by these criteria, “Charles” and “James” are strong candidates. Dr. Davidson was also taken with the Korean use of “Jay,” but she was concerned that it is too confusable with “Jake” in some settings. After some discussion, we settled on an option that we all agreed fit the criteria, and I resolved to order a coffee using the name.
I am on line, as the locals say, at the Starbucks on Astor Place, one of the great people-watching spots in lower Manhattan. I am a little nervous. This will be my first time using a coffee pseudonym, a coffeenym. It is not my first attempt. On my previous stop for coffee, I was never asked for my name. (The name on my coffee cup is always a problem—some of the time.)
The line inches forward. I have a plan: I will order an espresso drink to ensure that the cup goes to the barista with a name on it, and I will use the pseudonym that the NYU linguists and I chose together. Although it is after 9:00 in the morning, the place is still busy enough to warrant an associate standing in the middle of the line, receiving orders before we reach the registers. He greets me. I nod and I ask for a small latte. “Tall latte,” he directs into his walkie-talkie; then to me, “What’s your name?” By force of habit, I almost reply “Greg,” but catch myself just in time. I search his face as I give him the pseudonym, trying not to make it sound like a question. Unblinking, he repeats my order and name into the two-way.
The line moves swiftly to the register, where they don’t need to know my name. I pay, and circle around the espresso machine, angling for a vantage point from which I can clearly see and hear the barista. She sets a series of iced coffees atop the counter before giving a small white cup a quarter-turn to read the looping cursive.
She hesitates ever so slightly, then calls out, “Elvis!”
Elvis. One successful trip to the coffee shop doesn’t tell me much about the long-term viability of a pseudonym, but when Sean, the NYU grad student, suggested Elvis, it quickly became the favorite in the room. Just what is it about Elvis that makes it “the King” of coffee-cup names?
It ends with the sibilant /s/, a consonant of strength in an especially vulnerable position. It may not be the most identifiable name, but there are few names like it, at least to the ear of this American. “Elvin” may be the closest. Other similar names—“Olvis,” “Alvis,” or the Norwegian comedy act Ylvis—are so rare, they seem unlikely to register as actual names to the typical American coffee-shop employee.
Two more advantages—a well-known standardized spelling and cross-cultural recognition as a name—are the result of Elvis Presley’s fame. On the other hand, celebrity names are ripe for scrutiny, and an obviously fake name runs the risk of resentment from the coffeehouse staff, who might feel like the target of a silly joke. In this case, the benefits of celebrity seem to warrant the cost to credulity.
Someday the enormity of the coffeenym problem might be properly recognized, resulting in grant funding for exhaustive fieldwork and a complex data-driven approach. For now, I must be satisfied with Elvis and its theoretical superiority.
Although I’ve completed my quest for a coffee-cup pseudonym, I feel a lingering emptiness. Maybe this is because I set out to find a phonetically optimal universal, but had to settle for something more like a social lubricant that must be judiciously deployed—not around the corner, and not at any of my regular haunts, but only in places where intelligibility benefits me more than I lose through the artifice.