Words on the Move

by John Mcwhorter

  You’d never guess, but it made sense. Think of a southerner with a good solid drawl saying meet—it’s more like “muh-eet.” It isn’t hard to start pronouncing a vowel with a little uh before it—note today how often younger people say “Wuh-it???” rather than “What?” From meet to “muh-eet” was the exact same pathway, for the same reason of articulatory serendipity. Little alternate ways of saying things, among the infinite array of ones possible, settle randomly in.

  But look: in our chart, that little uh in “muh-eet,” uh being the but sound, sits a step up from ah down on the lower right. Hence, over time “muh-eet” became “mah-eet”—that is, our mite. In terms of where it is pronounced in the mouth, the vowel in mite is neither fish nor fowl: one part ah and one part ee, it doesn’t fit neatly above beet in any coherent way. Rather, mite just branched off in a little spew:

  Then, what happened to oo when moved in on by words like “mohhs”? Same thing: the Old English rendition of mouse was in the oo slot, pronounced “moose.” Now, a southern American transplanted to Merrie Olde England might pronounce it as “muh-oos” in his accent. As it happened, a Middle English speaker was given to saying it just that way. First “muh-oos,” eventually a step southeast to “mah-oos,” and that’s mouse. (Then, many speakers move that ah in “mah-oos” over a notch leftward to the bat sound.) So what began as an imposing car-smashing horned mose and a squeaky little rodent moose became a horned moose and a squeaky little mouse. The vowel in mouse no more fits in above boot in our chart than bite fits in above beet, and so we can draw it in as a complementary spew on the right:

  There was actually one last little afterpiece, which finally explains the good/food problem. Later on, some words that had wound up in the boot slot made another move, down left into the book slot. Food, as it happened, stayed put, but good moved to book—voilà, the ruin of what once rhymed in the famous prayer. Then some words like this moved even farther in, to the but slot—such as blood and flood. This is why, then, food, good, and flood don’t rhyme. Hence also the trio poot, put, and putt—given that u in languages usually means the oo sound, we can see how all three of these words “should” be pronounced “poot,” and they once were.

  All that was vowel shifting, of the exact same kind that makes a person in Detroit talk about getting a “jab.” “Jab” looks somewhere between trivial and bemusing, but it is a modern symptom of what, occurring bit by bit and piling high, blocked us from being able to converse with Chaucer. The enshrinement of the Great Vowel Shift in the traditional story of the English language is like documenting the classical symphony of Mozart and Beethoven as the essence of Western music, as if Palestrina, Handel, and Bach were just preludes and Debussy, Schoenberg, Gershwin, and bebop were just footnotes. Just as music has always changed and we couldn’t imagine that it wouldn’t, vowels always shift and make words drift into new renditions of what they once temporarily were.

  One Never Knows

  The analogy with bees in the hive must be pushed further, to show how it applies to so much around us. For example, so far it may seem like vowel shifts are almost always clockwise for some reason. We’ve seen “mohhs” going counterclockwise to become moose, but seemingly only because the equivalent thing (“mate” becoming meet clockwise) was happening over on the other side. But the wheel goes both ways.

  In Old English, dogs chewed on “bahns,” it was sticks and “stahns” that broke your “bahns,” and there was no place like “hahm.” The ah sound rose, counterclockwise, and thus by Middle English the words were bone, stone, and home.

  Or, there is also a vowel shift happening in California. The bit vowel has collapsed, counterclockwise, down into the bet space, such that someone saying bit sounds more like they are saying bet. The mental domination of the written word, plus context, ensures that when a young Californian says “a little bet” for a little bit, we do not mishear. However, if you know any such people, play a mental audio file of them saying “leave a tip” and note how their ih sound has a lot of eh in it. Or, imagine, if I may, them saying the word for a female dog, and we have the explanation for the “Betch” joke. The routine stems from California, and “betch” is how women below a certain age in that state actually say bitch. The California shift is a choo-choo train chain, too: if bit is in bet, then bet likely moves on, and it goes down to the bat slot. “Make the bed,” a young or youngish someone from Santa Barbara says, and it sounds more like “Make the bad.”

  Thus vowel shifts can go either way, even at the same time. In California you have chains within chains, gears within gears within gears, shifting in opposite directions. People in California (and a ways eastward, especially Texas) often pronounce pink as “peenk” and king as “keeng.” It can seem like a little oddity—I’ve seen it mocked—but in terms of our chart it’s business as usual. Pink has the vowel in bit; “peenk” is the vowel in beet, just a step up. But wait—if this is California, why didn’t pink become “penk” like bit became bet? Because ng, n, and m have a way of creating their own little dramas. They are the nasal consonants—the ones you can buzz on—and making them requires closing the mouth at least somewhat at the back. When you say a word, your brain has the whole word planned before it comes out, and when we say a vowel, we have a way of anticipating what the sound after it is going to make the tongue do. When an m, n, or ng is coming, that means anticipating the tongue lift, such that pink goes up, clockwise, to “peenk.” And man, down below at bat, even comes up for the ride, to “mee-an”—the buzzing consonants can even pull that bat vowel all the way up to beet. When I did my graduate work in California, I noticed that people born there tended to say “mee-an” for man (and “stee-ind” for stand, “hee-um” for ham, and so on); linguistics taught me why.*

  Vowels really are, then, like bees in a hive, bumbling all over the place. You never know just where a bee might wiggle off to—or if chaos theory experts can tell us, the explanation would likely somehow not feel quite like one. Say you’re watching episodes of the old sitcom treat The Many Loves of Dobie Gillis (as all of us of course do from time to time) and in one of them the father character calls an orchid an “ar-chid,” asks, “What is this used far?,” and tells someone, “You’re getting ‘wahhrm’!” instead of “wore-m,” as most of us say the word. It’s a subtle thing, and so there’s no way the actor, Frank Faylen, was directed to use ar for or in this way. Rather, this was the way Faylen happened to talk, and it’s no accident that Faylen was from St. Louis. A peculiarity of pronunciation there is that, to quote the most cherished example, Highway 40 is pronounced “Highway Farty.” On the Gillis show, Faylen called corn “carn,” and so on.

  But it’s only so peculiar—the ah slot is just below the aw slot, and therefore “fawr-tee” to “farty” is a quick trip. Next time you’re with someone from St. Louis, listen for it—although they can’t be, well, young. The “wahrm archid” pronunciation is on its way out—it’s shining bright in the elderly, more modest but steady in their kids, and all but nonexistent among young people. For example, on the sitcom The Office (U.S. version), both Phyllis Smith (as Phyllis) and Ellie Kemper (as Erin, and later well-known as Kimmy Schmidt) are St. Louisans. Smith (sixtyish at the time of the show) had the ar for or; Kemper (then thirtyish) did not.

  But is this St. Louis “farty” thing weird? Only if we discount something happening across the nation. Take away the r in for and far, and the pronunciation of the vowels alone, aw and ah, is becoming the same for ever more people across America these days. As a linguistics teacher, I find it an ever-weirder challenge to teach what is becoming an ever-growing number of students in any class that, for some people, cot and caught do not rhyme. To me, cot is “caht” and caught is “cawt.” But that’s because I grew up in Philadelphia. For people in the western half of the country and most of New England as well as among many people in the Northern Cities Shift zone and more people elsewhere by the decade, both cot and caught are pronounced “caht.”

  Here, the
n, is why so many people spontaneously rhymed Dawn (“Dahn”), Sean (“Shahn”), John, and Ron. A vowel shift is happening, and has been for a long time. I recall taking issue with a girl in my elementary school class in the early seventies when we were talking about the television show Maude. (My mother made me watch it because it was educational about societal issues—and she just liked it—but yes, it is odd that I was talking about it with little Julie Pinchuk.) She pronounced it “Mod,” and I kept saying, “No, it’s Maww-d!” Either Julie or her parents were probably transplants from somewhere north or far west of Philadelphia, where this conversation took place. But Julie was a sign of the times: vowels shifting just as they had been during the Black Plague. They’ve never stopped.

  Thus what seems vocally odd is just vowels making their chess moves, despite how culturally rooted the shifts can accidentally seem. Brooklynites in the old days were well known for having “oy” for “er,” as in “toity-toid” for thirty-third. As a child I heard a glimmer of this from the last generation of young people to grow up with it—sometime in the early seventies, when my family took me on a trip to New York on the Fourth of July (I forget why), a kid of about nine or ten was on the sidewalk selling “Fire-woiks! Fire-woiks!” It didn’t occur to me until much later, however, that my own older black southern relatives had the exact same pronunciation of such words—shirt was “shoit,” work was indeed “woik.” Louis Armstrong is probably the most recorded black person exhibiting this vowel shift. On recordings he made in his home he casually talks about how his horn comes “foist” and how he needs someone to keep his horn “poi-colatin’” (for percolating).

  However, this was hardly unique to him, nor, as sometimes implied in sources about oy for er, only a trait of the New Orleans where Armstrong happened to have grown up. It was normal in the English of the Deep South in the old days. One catches it regularly in black performers born across the South in the early twentieth century,* cases in which one might wonder why they have “Brooklyn” accents when actually it was just that the bees happened to make the same move where they grew up as they did in Brooklyn. After all, a hive is only so big.

  Then, finally, you never know how far those bees will skitter—they can stop shorter in some places than in others. For example, we saw how in what became Standard English, vowel-wise some words jumped from boot to book and then to but: what was once blude jumped two steps in to bludd. But in northern England and Scotland, words did not take the second step to the but slot, and instead just stayed put in the put slot. This is why people speaking those dialects say but, love, and blood with a vowel that rhymes with book instead of but. Most of us just hear it as quaint or as what is typical of the non-elite characters in Game of Thrones. But it’s actually quite predictable from the layout of our grid, otherwise known as the human mouth.

  Also, take the grand olde Scottish folk song “Annie Laurie”—a singer doing a savory rendition will end it with “I’d lay doon my head and dee” for I’d lay down my head and die. Why do the Scots say it “like that”? There’s an easy reason—there was no “spew” in Scotland! In what became Standard English, down started as “doon” and then became “down” just as “moos” became mouse and “hoos” became house. But in Scotland, words like “doon” just stayed where they were—no spew into “down.” And there was a systematicity to this, in that over in front, the words up in ee didn’t spew into “aye,” either. Just as our bite started as pronounced “beet,” our die started as pronounced “dee.” In Scotland it stayed that way—hence lying doon to dee. One never knows—at least one can’t beforehand, but one can nevertheless understand afterward.

  Oh, yes—what about Jack and Jill? Well, it’s one more lesson in how sounds change in predictable ways that leave strange-looking results. You’re British and you say AHF-ter. Said for eons, and often, as is a word like after, you can imagine that it could be shortened into AH-ter—which, note, rhymes just fine with water! In many dialects of English in Britain, after actually has long been “ah-ter,” and “Jack and Jill” was written in one of those dialects. That doesn’t really follow directly from the particular things I have discussed in this chapter, but I have always wanted to get this in somewhere, and you must admit it’s fun to know.

  How Should I Pronounce?

  With vowels this restless even today, it follows that people even as recently as a hundred or so years ago would sound odd to us simply because their vowels, like their hair styles and hemlines, were in different places than ours today. We can in fact know this, from what people at the time confidently proclaimed as the “proper” way to pronounce words.

  A perfect example was one William Henry P. Phyfe, an American specialist in pronunciation in the late nineteenth century, who made a career with books teaching people the “best” way to speak. Anyone who followed his advice in our times would be at a distinct social disadvantage, in pronouncing words in ways we would consider quite beyond consideration. Often it’s because vowels can’t stay put.

  “Correct pronunciation,” Phyfe assured us in the preface to one of his books, How Should I Pronounce?,

  is the best prima-facie evidence of general culture. On this account it appeals to all, since there is no one wholly indifferent to the estimate formed of his social position, and who, in consequence, would not cultivate those arts that are at once the criteria of social standing and the stepping-stones to a more liberal culture.

  The man was serious. It would be too easy to mock him as a clueless social striver who didn’t understand that language is inherently changeable, though. He actually understood this better than many of his equivalents of the era:

  Some people, for instance, seem to think that the pronunciation of words never varies. Tell them that it is constantly changing; that there are many words that are pronounced in several different ways; that the question of correctness in this matter is one wholly of custom;—and we unfold to their minds a condition of things of which they before had no conception.

  Yet Phyfe’s idea is that the way “cultivated” people happen to be pronouncing words at his given moment is, for better or for worse, what anyone who seeks their status (station?) must imitate.

  That means that Phyfe’s little book allows us to listen in on how the kinds of people in Edith Wharton novels were pronouncing words in their drawing rooms, or at least the way they thought they were supposed to be pronouncing them. And, boy, would these people have sounded odd to us! Phyfe provides a lengthy glossary of words to watch out for, and we encounter one word after another for which he crisply prescribes a pronunciation that today would have people telling us we probably need to get more sleep.

  Apparently there were people in 1885 pronouncing daunt, haunt, and taunt as “dahnt,” “hahnt,” and “tahnt,” and Phyfe thinks of these as “correct.” That people would pronounce these words that way is not surprising in itself, given that, as we have seen, the baht vowel is just a step away from the bought vowel that so many of us pronounce daunt, haunt, and taunt with today. But Phyfe does not overall suggest that all aws be pronounced as ah, as in today’s California (and elsewhere) accent, with cot for caught, hock for hawk, and rah for raw. That accent didn’t exist in 1885. What Phyfe was advising was something much more specific: aw becoming ah before just n. As Gilded Age sophisticates, we were also to say “lahnch” for launch, “lahndry” for laundry, and of course, “ahnt” rather than “ant” for aunt. Then also, apparently, for consistency’s sake, can’t was to be pronounced “cahnt” like the name of the philosopher.

  During the administrations of presidents like Chester Alan Arthur, Grover Cleveland, and Benjamin Harrison, then, the cultivated American might correct herself into saying “lahndry” instead of laundry and tell her children to do the same, and talk of “hahnted” houses. In 1885: music in different forms, clothing in different cuts, vowels in different slots. For Phyfe, isolate was “izzolate,” equable was “ee-quable,” sacrilegious was “sacriLEEjus,” and nomad was “NAH-mad.”
A person of grace and substance just 125 years ago could say, in all seriousness, “It would be sacrileejus to izzolate those nahmads.” Things change, for no real reason. Across the Atlantic, in France, about fifteen years after Phyfe published his book, two British academics were touring Versailles when, they claimed, they slipped into a time warp back to 1789. They later said they spoke to various people, including Marie Antoinette. You are not alone in finding this story unlikely; the two women were politely ridiculed for years after they wrote a book on the subject. However, there is one thing in their account that does ring true: they mentioned that the French they heard was oddly pronounced. Not just “olde” in terms of vocabulary and grammar (although they mentioned that, too), but because people sounded funny. That detail isn’t something most people, academics or not, would know to include in an account like theirs about just a century-and-change before. “Funny” is exactly how the French of 1789 would sound to modern speakers, because sounds ooze around—a’s then weren’t quite like a’s now, etc. We can know the same thing about American English at the same remove from us as 1789 was from those professors, with a look at books like Phyfe’s.

  Consonants get jostled around as a language moves along, too, and Phyfe’s American English demonstrates this amply. As with vowels, consonants don’t differ in random ways; what Phyfe wants is always a close cousin of the consonant we use today, such as soft versions of consonants for which we are accustomed to the hard versions.* Dishonest is to be pronounced “diz-honest,” suffice “suffize,” and Greenwich “grinage.” Now and then Phyfe shows us, in what he shoos away, things that are just plain odd. At girl, he advises, “Do not introduce a y sound before the ‘i’ in this word. It is regarded as an affectation.” Of course, today, none of us need be told not to say “gyirl.” Who would? Phyfe’s comment shows that in 1885 there were people actually going around saying “gyirl” and thinking of it as classy! Who knew?